Abstract
Futile recanalization (FR) after endovascular treatment (EVT) remains a significant challenge for acute ischemic stroke (AIS) with large vessel occlusion (LVO). The pathogenesis of FR has not been well elucidated. We prospectively enrolled anterior circulation LVO-AIS patients who achieved successful recanalization after EVT. The jugular venous blood ipsilateral to stroke was collected before and immediately after recanalization. Plasma proteomic analysis based on liquid chromatography-mass spectrometry was performed using data-independent acquisition method. Differentially expressed proteins (DEPs) among patients with or without FR in the whole or propensity score matching (PSM) cohorts were screened according to the absolute value of fold change ≥1.5 and P value <0.05. We identified 104 and 34 DEPs between patients with or without FR in the whole cohort and PSM cohort, respectively. Bioinformatic analysis indicated that the identified proteins were primarily related to specific biological processes including immune response, complement activation, oxidative stress, lipid metabolism, protein ubiquitylation as well as autophagy, suggesting that these may be mechanisms in FR pathogenesis. Collectively, we discovered proteins that may be potential research targets for FR. The combination of proteomic and bioinformatic analysis could provide a better understanding of the pathogenesis of FR in a comprehensive manner.
Keywords
Introduction
Decades of concerted effort have enabled progress and improvement in the field of reperfusion therapy for acute ischemic stroke (AIS).1,2 Evidence from several randomized controlled clinical trials have established that endovascular treatment (EVT) is of benefit to most patients with AIS caused by large vessel occlusion (LVO) of the anterior circulation within 6 hours after onset3 –5 or 6–24 hours with additional imaging criteria.6,7 However, despite successful recanalization, approximately half of the patients remain with 3-month functional dependency (modified Rankin Scale [mRS] of 3–6), which has been termed futile recanalization (FR)8,9 or reperfusion without functional independence. 10 Uncovering the pathophysiology or mechanisms of FR may help to optimize adjunct therapeutic strategies.
The relationship of blood biomarkers as a practical tool to the pathophysiological mechanisms of FR has not been fully investigated. Several studies found that neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, matrix metalloproteinase-9, tenascin-C, thioredoxin, ADAMTS13 and gelsolin were independently associated with FR. 11 To our knowledge, there have been no comprehensive studies to investigate the association of changes in blood biomarkers before or after recanalization with functional outcomes in AIS patients caused by LVO and received EVT, which could be helpful in understanding the mechanisms underlying FR.
Considering that the jugular vein drains the cerebral circulation before it enters into the pulmonary circulation or its tributaries, analysis of ipsilateral jugular venous blood before and immediately after recanalization could provide a more tailored understanding of the local pathological microenvironment. In this context, we prospectively collected jugular venous blood before and immediately after successful EVT recanalization in anterior circulation LVO-AIS patients and performed a comparative proteomic analysis of plasma samples using nano liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) method. The results of this proteome study may help provide insight into the mechanisms of FR and may also provide knowledge about FR proteome biomarkers.
Materials and methods
Study design
Patients with AIS due to anterior circulation LVO were screened for eligibility for the study at Dalian Municipal Central Hospital between January 2022 to December 2022. The study was conducted according to the World Medical Association Declaration of Helsinki and approved by the ethics committee of Dalian Municipal Central Hospital (YN2022-096-13). Informed consent was obtained from all patients, or their families if the patient had functional limb paralysis or loss of consciousness. The inclusion criteria were as follows: (1) AIS patients who underwent EVT for anterior circulation LVO (intracranial terminal of internal carotid artery [ICA] and/or M1 segment of middle cerebral artery [MCA]) confirmed by pre-treatment CT-angiogram and digital subtraction angiography (DSA); (2) age ≥18 years; (3) initial National Institute of Health Stroke Scale (NIHSS) ≥6; (4) informed consent was obtained from patients or their families.
The exclusion criteria were: (1) pre-stroke mRS ≥2; (2) intracranial hemorrhage or intracranial tumor on admission; (3) preexisting neurological or psychiatric disease that would confound neurological evaluation; (4) active or recent hemorrhage, or history of surgery within 2 months before stroke onset; (5) concurrent infection at the time of sample collection; (6) rheumatoid immune diseases, severe liver/kidney diseases, hematopathy, or malignant tumors; (7) infectious embolus or bacterial endocarditis; (8) baseline platelet count <100 × 109/L; (9) failure of recanalization (modified Thrombolysis in Cerebral Infarction [mTICI] grade 0–2a after EVT); (10) loss to follow-up; (11) other conditions unsuitable for this study as determined by investigator.
Clinical data
The clinical characteristics and information including age, sex, weight, height, history of hypertension, diabetes mellitus, hyperlipemia, atrial fibrillation, coronary artery disease, smoking, alcohol consumption, initial NIHSS score, treatment with intravenous thrombolysis, time from onset to puncture/recanalization, occlusion artery, number of passes to achieve recanalization, laboratory parameters including fasting blood glucose and creatinine, mRS score at 90 days after EVT were prospectively collected. Stroke subtypes were determined by the Trial of Org 10,172 in Acute Stroke Treatment (TOAST) classification. Baseline Alberta Stroke Program Early CT Score (ASPECTS) was measured based on CT before treatment. FR was defined as poor outcome (mRS ≥3 at 90 days) in patients who achieved successful recanalization (mTICI ≥2b) after EVT. 12 According to the presence or absence of FR, all patients were divided into two groups: FR group and non-FR group. Patient loss of follow-up was defined as being out of contact and unable to obtain mRS score at 90 days after stroke onset, and these patients were further excluded in this study.
Plasma collection and proteomic analysis
Blood sample from the jugular vein of each enrolled patient was collected. Briefly, a catheter was placed in the ipsilateral internal jugular vein via transfemoral venous puncture before the procedure and until the procedure was done. A syringe was connected to the external end of catheter, 5 ml venous blood was slowly drawn out before (defined as PreV) and immediately after (defined as PostV) occluded artery recanalization, and then quickly transferred to an EDTA anticoagulation tube. After centrifugation (1200 g, 4°C) for 10 minutes, the supernatant was aliquoted and cryopreserved in liquid nitrogen before being transferred to a −80°C fridge for storage until assayed.
Proteomic analysis using the LC-MS/MS method was performed by BGI (Shenzhen, China) as previously described. 13 The Q-Exactive HF X (Thermo Fisher Scientific, San Jose, CA) was used to acquire MS data for 80 samples from 40 patients (20 FR patients and 20 non-FR patients, PreV and PostV samples from each patient) in data-independent acquisition (DIA) mode. Differences of plasma proteome were statistically evaluated by the MSstats software package.
Bioinformatic analysis
The partial least-squares-discriminant analysis (PLS-DA), a classic PLS regression for solving discrimination and classification problems, was performed by the mixOmics package. Hierarchical clustering analysis was performed using the pheatmap package. The volcano plot was generated by the ggplot2 package for visualization. Functional annotation, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed by cluster Profiler package. A Search Tool for the Retrieval of Interacting Genes (STRING, 11.5) software (www.string-db.org) was used to analyze the protein-protein interaction (PPI) network. The above analyses are important to understand cell physiology, function and mechanisms under normal and diseased conditions.
Propensity score matching
To exclude the potential effect of baseline and clinical characteristics, propensity score matching (PSM) was performed between the FR and control participants with the ratio 1:1, and operated with age, gender, diabetes mellitus, hypertension, and ASPECTS which were common factors associated with stroke prognosis.
Statistical analysis
The selection of patients for proteomic analysis was performed with a completely randomized method (drawing lots), data analysis was conducted with a blinded fashion. The differential expressed proteins (DEPs) were screened out according to the absolute value of fold change (|FC|) ≥1.5 and P value <0.05 as the judgment criteria for significant differences. Protein intensity was defined as the arithmetic mean of the peptide peak intensities attributed to the protein. All undetectable proteins were recorded as zero. Since this is an exploratory study, only 20 patients were randomly selected from each group. Continuous data distribution was examined by the Shapiro-Wilk test. Normal distribution data were presented as mean ± standard deviation (SD). The intergroup differences between the two groups were assessed using independent t-test. Data without normal distribution were presented as median with interquartile range (IQR). The Mann-Whitney test was used when comparing two groups. Categorical variables were presented as frequency rates and percentages and were analyzed using the chi-squared test or Fisher’s exact test, as appropriate. Statistical analysis was performed using SPSS software package (Version 22.0, USA), MedCalc (19.0.7), and R software package (3.5.2). P < 0.05 was considered to be statistically significant.
Data availability
The raw data supporting the findings of this study are available from the corresponding author (Hui-Sheng Chen,
Results
Clinical characteristics
The total follow-up time for each patient was 90 days. After excluding eight patients (three with poor blood sample quality, three with failure of recanalization [mTICI < 2 b] after EVT, two with loss of visit during follow-up), 48 eligible AIS patients with successful recanalization after EVT fulfilled the criteria for enrollment in our study, including 23 with FR and 25 without FR (please refer to Figure 1 for flow diagram). Among them, 20 patients from each group were randomly selected for further proteomic analysis. The baseline characteristics between patients with or without FR are described in Table 1. The age of patients in the overall population was 67.10 ± 11.28 years, with men accounting for 80%. Patients with FR were older (73.15 ± 8.03 vs 61.05 ± 10.93), had higher fasting blood glucose [9.38 (7.09–11.22) vs 7.06 (6.15–8.85)], more ICA occlusion (45% vs 10%), hypertension (75% vs 25%) and diabetes mellitus (50% vs 15%). No significant differences between other characteristics were found.

The flow diagram of the study.
Clinical characteristics of AIS patients with or without futile recanalization.
P values in bold font represent those with statistically significant differences.
BMI: body mass index; FBG: fasting blood glucose; NIHSS: National Institute of Health Stroke Scale; ASPECTS: Alberta Stroke Program Early CT Score; TOAST: Trial of Org 10:172 in Acute Stroke Treatment; MCA: middle cerebral artery; ICA: internal carotid artery; ORT: onset to recanalization time; OPT: onset to puncture time.
Identification of the DEPs between patients with or without FR
Details on the proteomics including peptide numbers and sequences are listed in Additional file 1, protein scores and levels are listed in Additional file 2. A total of 2020 different proteins were identified between groups. The basic statistical results of the plasma proteins are shown in Supplemental Figure S1, which demonstrated that 58.32% of proteins had a number of unique peptides ≥3, most of these proteins were medium proteins (molecular masses 10–60 kDa) and the protein coverage distribution was mainly concentrated within 60%.
The plasma proteome differences between patients with or without FR were compared to investigate the significant DEPs before (PreV) and after EVT (PostV). The PLS-DA model showed some differences between groups, especially in PostV samples (Figure 2(a)). There were 26 and 63 functional proteins exclusively dysregulated before EVT and after EVT, respectively. Fifteen functional proteins were dysregulated both in PreV and PostV plasmas between patients with or without FR (all P < 0.05; all DEPs and their functions are listed in supplemental Table S1 and Table S2).

The PLS-DA model, volcano plots and heat map of differential proteins of the FR and non-FR (NFR) groups. (a) The PLS-DA model of differential proteins in PreV and PostV plasma samples. Partial separation of the two groups was observed. (b) The volcano plots of the differential proteins in PreV and PostV plasma samples. Red dots in the left X axis (log1.5 (fold change) <-1) represent up-regulated proteins in FR group, in the right X axis (log1.5 (fold change)>1) represent down-regulated proteins in FR group, and dots in gray represent proteins without obvious significant. (c) Heatmap of the differential proteins in PreV and PostV plasma samples.
Volcano plots revealed 27 up-regulated and 14 down-regulated functional proteins in PreV samples of FR group (Figure 2(b)). The top five up-regulated proteins (ranked by |FC|) were proteasome subunit alpha type-6, Mth938 domain-containing protein, salivary acidic proline-rich phosphoprotein 1/2, D-dopachrome tautomerase and adenosine deaminase 2 with a fold change of 8.46, 7.20, 6.66, 6.58, 5.54, respectively. The top five down-regulated proteins were 5′(3′)-deoxyribonucleotidase, serpin B9, Golgi reassembly-stacking protein 2, NIF3-like protein 1 and peptidyl-prolyl cis-trans isomerase FKBP3 with a fold change of 6.464, 6.129, 4.642, 4.120 and 3.902, respectively (supplemental Table S1).
In the PostV of FR individuals, it contained 49 up-regulated and 29 down-regulated functional proteins compared with the non-FR group (Figure 2(b)). The top five up-regulated functional proteins were histone H1.3, ADP-ribosylation factor-like protein 8A, C-type mannose receptor 2, proSAAS and elafin with a fold change of 8.728, 7.834, 7.489, 6.549, 6.436, respectively. The top five down-regulated proteins were Golgi reassembly-stacking protein 2, cytochrome b5, proteasome subunit alpha type-5, NADH dehydrogenase [ubiquinone] 1 alpha subcomplex subunit 9 and protein canopy homolog 2 with a fold change of 8.896, 7.097, 5.758, 4.996, 4.684, respectively (supplemental Table S2). Heat map analysis exhibited levels of DEPs, providing information on the underlying disturbance caused by FR (Figure 2(c)). These results suggested that the patterns of differential protein expression varied between FR and non-FR groups.
GO annotation and enrichment analysis
The GO enrichment analysis included three categories: cellular component, molecular function and biological process. For the cellular component category, the DEPs were mostly enriched in immunoglobulin complex, haptoglobin-hemoglobin complex, hemoglobin complex in PreV samples, and in phagocytic cup, primary lysosome, azurophil granule in PostV samples (Figure 3(a)). For the molecular function category, the DEPs were predominantly involved in peroxidase activity, oxidoreductase activity, and antioxidant activity in PreV samples and in complement component C1q complex binding, opsonin binding and complement binding in PostV samples (Figure 3(b)). For the biological process category, DEPs were associated with the response to cellular detoxification, cellular response to toxic substance, detoxfication in PreV samples and response to phagocytosis, complement activation and plasma membrane invagination in PostV samples (Figure 3(c)).

GO analysis of the differential proteins in FR and control individuals. (a) Cellular components in PreV and PostV samples. (b) Molecular functions in PreV and PostV samples and (c) Biological processes in PreV and PostV samples.
KEGG enrichment analysis
The biological function of identified DEPs was further characterized via KEGG enrichment analysis. The top five significant enrichment were dilated cardiomyopathy, nucleotide metabolism, hypertrophic cardiomyopathy, amoebiasis, and purine metabolism in PreV samples (Figure 4(a)), as well as staphylococcus aureus infection, phagosome, drug metabolism other enzymes, systemic lupus erythematosus, and tuberculosis in PostV samples (Figure 4(b)). After combining the PreV and PostV DEPs, the ten highest ranked biological functions were phagosome, neutrophil extracellular trap formation, transcriptional misregulation in cancer, asthma, african typansomiasis, allograft rejection, primary immunodeficiency, intestinal immune network for IgA production, autoimmune thyroid disease, and the notch signaling pathway (Figure 4(c)).

KEGG pathway enrichment analysis. The KEGG pathway enrichment analysis for DEPs from the PreV (a), PostV (b), and combination of PreV with PostV (c) plasma samples.
PPI network of DEPs
The PPI network was constructed to illustrate the interactions among DEPs. Thirty-nine of the 41 proteins from PreV samples were matched in the database. The 5′(3′)-deoxyribonucleotidase (NT5C) and myeloperoxidase (MPO) seemed to be the potential key proteins (supplemental Figure S2). Meanwhile, 76 proteins from PostV samples were matched in the database, MPO, low affinity immunoglobulin gamma Fc region receptor III-A (FCGR3A), elongation factor 1-beta (EEF1B2), eukaryotic peptide chain release factor subunit 1 (ETF1), T-complex protein 1 subunit delta (CCT4), isocitrate dehydrogenase [NADP] cytoplasmic (IDH2), complement component C1q receptor (CD93), syndecan-1 (SDC1) and pyruvate kinase PKM (PKM) were considered to be the potential hub targets. Notably, MPO connecting seven nodes was considered as the most significant (supplemental Figure S3).
DEPs in propensity score matching cohort
We included age, history of diabetes mellitus and hypertension which were significant differences between patients with or without FR as well as gender and ASPECTS which consider to be the important parameters in stroke prognosis for propensity score matching. A total of 12 patients were matched based on propensity scores (n = 6/group). Demographic characteristics and clinical data of the two populations were similar (supplemental Table S3). In this cohort, we identified 34 DEPs between FR and non-FR individuals (all listed in supplemental Table S4 and Table S5), including 11 exclusively dysregulated in PreV, 21 exclusively dysregulated in PostV and two dysregulated both in PreV and PostV samples. Moreover, expression of proteasome subunit alpha type-6, mth938 domain-containing protein, carbonic anhydrase-related protein 11, glutathione S-transferase Mu 1, glutathione S-transferase Mu 5, Ras-related protein Rab-4A and WAP four-disulfide core domain protein 2 which involved in biological processes including lipid metabolism, 14 oxidative stress, 15 membrane vesicle transport and others 16 were both upregulated in FR patients from the whole population (n = 20/group) and the propensity score matching population (n = 6/group). PPI network demonstrated that most of these proteins were independent of each other.
Discussion
In this study, we used proteomic methodologies (LC-MS/MS) to systematically investigate the proteomes of plasma collected from the jugular vein of AIS patients before and after EVT. By comparing the differential proteins, we identified 41 and 78 important DEPs before and after recanalization between patients with or without FR in the whole population. The role of most of these proteins in stroke has rarely been investigated and we provide evidence of these proteins in connection with stroke in this study. We found that these proteins mainly participate in protein degradation and folding, adipogenesis or atherosclerosis, inflammatory response, maintenance of blood-brain barrier integrity, angiogenesis and intercellular adhesion, autophagy, neuroendocrine secretory, electrolyte homeostasis, vesicular traffic, glucose metabolism, redox reaction, as well as cell growth/proliferation, migration or apoptosis (details are listed in Table S1 and S2). In the PSM population, we identified 13 and 23 differential proteins in the jugular vein before and after recanalization between the FR and non-FR groups, respectively. Most of them were upregulated in FR individuals and also have rarely been studied in stroke before (further details in Table S4 and S5).
A difference protein profiles before and after EVT was observed in the jugular vein plasma. Based on the GO analysis, we found an abnormality mainly associated with oxidative stress before EVT, whereas after EVT, the abnormality was primarily related to immune regulation. Besides, the ubiquitin-proteasome system (UPS) is expected to play a role in the onset and progression of stroke via multiple targets and pathways, 17 and is closely related to the pathways of post-stroke related pathological changes such as mitochondrial autophagy, oxidative stress, hypoxia and inflammatory response. 18 We found that the expression of two proteasome core subunits, proteasome subunit alpha type-6 (PreV, FR up-regulated) and proteasome subunit alpha type-5 (PostV, FR down-regulated), which belong to component of the 20S core proteasome complex that participates in the degradation of ubiquitinated proteins, were dysregulated inconsistently. We surmise that the increase of proteasome subunit alpha type-6 expression before EVT is to promote protein ubiquitination and degradation, however, the decrease of proteasome subunit alpha type-5 expression after EVT leads to the accumulation of misfolded or harmful proteins thus facilitating the occurrence of FR.
Our study differs from previous work in the following ways: 1) Here, we compare the differential protein profiles before and after EVT via LC-MS/MS. This high-throughput proteomics method provides information on the expression of numerous proteins at a specific stage. 19 By comparing the proteomes of normal and pathological individuals, it is possible to identify certain “disease-specific protein molecules” that might serve as molecular targets for the design of new drugs or provide molecular markers for the early diagnosis of diseases. 2) We explore the protein changes immediately after vascular recanalization. Although several studies have characterized the plasma response to stroke before 20 or in the hours and days11,21 after EVT, the immediate effect after successful recanalization of diseased vessels of cerebral ischemia has not been assessed. 3) We collected jugular venous blood from EVT patients for detection and analysis. As a downstream draining vein capable of rapidly receiving signals released from the ipsilateral occluded intracranial artery before and after recanalization, a jugular venous sample could provide a more comprehensive description of the local pathological microenvironment. In light of the above advantages, we expand the currently available evidence about the pathogenesis of FR after EVT in acute large vessel occlusion ischemic stroke patients.
Our study has several limitations. First, the sample size (n = 6/group) is small after propensity score matching which may not accurately represent all FR patients. Collecting blood through a jugular vein was not part of a routine endovascular treatment procedure and getting informed consent from patients/relatives before surgery was also challenging, which greatly limited the number of patients enrolled. Though small sample size may hinder the interpretation of neutral results, more than 30 proteins found to have a significant difference from our investigations may indicate a difference between the FR and control groups. Nevertheless, larger populations with balanced clinical characteristics are imperative to verify these results. Second, blood samples from the ipsilateral jugular vein could provide certain information about a specific stage in FR progression. The content of proteins that can be detected in peripheral venous blood may decrease after passage through the systemic arteriovenous circulation. Further study to confirm the potential biomarkers in peripheral venous blood and its correlation with jugular venous blood is recommended. Third, only two time points, before and immediately after recanalization, were considered in this study. Future research with serial measurements could provide an opportunity to explore whether the presence of dysregulated target proteins persists at later time intervals.
In conclusion, to our knowledge, this is the first study to report the proteomic alterations in the jugular vein blood sample of patients with FR, which will provide valuable knowledge about plasma proteome characteristics of FR. The activation of inflammation, oxidative stress, autophagy, the disturbance of lipid metabolism and the ubiquitin-proteasome system might play an important role in FR, which may provide potentially novel therapeutic targets for FR.
Supplemental Material
sj-pdf-1-jcb-10.1177_0271678X231216767 - Supplemental material for Proteomic analysis of jugular venous blood in acute large vessel occlusion stroke with futile recanalization
Supplemental material, sj-pdf-1-jcb-10.1177_0271678X231216767 for Proteomic analysis of jugular venous blood in acute large vessel occlusion stroke with futile recanalization by Xiao-Yan Lan, Di Li, Yu Cui, Thanh N Nguyen, Shen Li and Hui-Sheng Chen in Journal of Cerebral Blood Flow & Metabolism
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from the National Natural Science Foundation of China (8207147, 82201626), Science and Technology Planning Project of Liaoning Province (2022JH2/101500020), Natural Science Foundation of Liaoning Province (2022-MS-442), Medical Science Research Planning Program of Dalian (2211007), Science and Technology project of the Summit Plan of Dalian (2022ZZ203).
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: T. Nguyen reports advisory board with Idorsia.
Authors’ contributions
All authors contributed to the study conception and design. XYL: experiment conduction, data analysis, data interpretation, manuscript writing, financial support. DL: blood collection, critical review of the manuscript; YC: data analysis and interpretation, critical review of the manuscript; TNN: critical review of the manuscript; SL: data analysis and interpretation, critical review of the manuscript; HSC: data interpretation, manuscript revision, financial support. All authors have read and approved the final version of the manuscript.
Supplementary material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
