Investigation of the relationship between findings detected in cranial MRI SWI and FLAIR sequences and infarct area and clinical prognosis in patients diagnosed with acute middle cerebral artery infarction

Authors

Abstract

Aim This study aimed to evaluate the prognostic value of combined Susceptibility Weighted Imaging (SWI) and Fluid-Attenuated Inversion Recovery (FLAIR) sequences in acute middle cerebral artery (MCA) infarction, focusing on their correlation with infarct area and clinical outcomes.
Materials and Methods A retrospective analysis was conducted on 44 patients diagnosed with acute MCA infarction who underwent MRI (SWI, FLAIR, and TOF-MRA sequences) between January 2016 and January 2017. Imaging findings, including susceptibility vessel sign (SVS) on SWI and hyperintense vessel sign (HVS) on FLAIR, were analyzed alongside infarct volume (ASPECT score) and clinical outcomes (NIHSS scores). Statistical analyses were performed using SPSS 15.0.
Results The co-presence of SWI SVS and FLAIR HVS was associated with significant NIHSS score improvement (ΔNIHSS: 3.20 vs. 0.88; p = 0.026). Thrombus location influenced outcomes, with MCA M2 occlusions showing larger infarct volumes but better recovery compared to ICA/MCA occlusions (p = 0.007). FLAIR HVS alone predicted neurological improvement (p = 0.043), while neither marker correlated with infarct expansion.
Discussion Combined SWI and FLAIR findings serve as effective prognostic biomarkers when used together, identifying patients with favorable collateral circulation and clinical recovery. The moderate sensitivity of SWI SVS (52%) is enhanced through combination with FLAIR HVS, demonstrating the value of multimodal assessment. Thrombus location significantly impacts outcomes, underscoring the need for tailored therapeutic strategies.

Keywords

Introduction

Stroke is one of the leading causes of mortality and morbidity worldwide. The World Health Organization defines stroke as a clinical condition of vascular origin, without any apparent cause other than vascular, characterized by rapidly developing focal or global cerebral dysfunction lasting 24 hours or longer or leading to death [1]. Today, particularly with the aging population, the incidence of stroke is increasing, resulting in significant socioeconomic consequences at both individual and societal levels. According to 2019 data, it was reported that 101.5 million people worldwide had experienced a stroke, with 77.2 million of these cases being acute ischemic stroke [2]. In the population over 60 years of age, cerebrovascular diseases are the second most common cause of death after cardiovascular diseases; moreover, stroke is the most frequent cause of disability and loss of workforce globally [2, 3] Ischemic stroke is a condition that causes sudden and irreversible damage to the brain parenchyma. Treatment success largely depends on rapid and accurate diagnosis. In the hyperacute phase (first 3–6 hours), computed tomography (CT) is useful for ruling out hemorrhage, but its sensitivity in detecting ischemic changes is limited [4, 5]. MRI, being more sensitive to changes in brain water content, is superior to CT in detecting acute infarcts. With MRI, up to 80% of infarcts within the first 24 hours can be detected. The MRI findings of cerebral infarcts develop over time, similar to CT [4, 6, 7]. Therefore, magnetic resonance imaging (MRI), particularly diffusion weighted imaging (DWI), has become the gold standard for diagnosing ischemic stroke [6, 7]. Additionally, perfusion MRI is used to assess cerebral perfusion, and 3D Time of Flight (TOF) MR angiography is employed to detect large vessel occlusions. Furthermore, FLAIR (Fluid-Attenuated Inversion Recovery), T1-weighted, and T2-weighted sequences are included in the imaging protocol to evaluate the temporal progression of the ischemic area [1, 8].
Studies have shown that MRI sequences not only provide primary diagnostic information but also offer additional insights that may correlate with clinical outcomes and guide clinicians in treatment and follow-up [4, 6, 7](Kim et al., 2014; Lansberg et al., 2017; Hasan et al., 2018). Specifically, FLAIR and Susceptibility Weighted Imaging (SWI) sequences have been demonstrated to not only identify lesions but also provide prognostic information regarding parameters such as intravascular thrombus presence, microcirculation disturbances, and bleeding risk [9]. Hyperintensity in the FLAIR sequence may indicate thrombus, while the “susceptibility vessel sign” in the SWI sequence, caused by magnetic susceptibility differences due to thrombus, allows non-invasive detection of large vessel occlusions [4].
In acute stroke, the findings detected in MRI examinations may provide significant information about the patient’s clinical outcomes, guiding treatment plans and contributing to the reduction of morbidity and mortality associated with stroke. The aim of this study is to investigate the relationship between findings detected in cranial MRI SWI and FLAIR sequences and the infarct area and clinical prognosis in patients diagnosed with acute MCA infarction.

Materials and Methods

This study was conducted as a retrospective investigation. The study was carried out using data from patients over 18 years of age, who were not pregnant, diagnosed with acute MCA, and underwent MRI examinations including SWI, FLAIR, and TOF MRA sequences at the Health Sciences University Şişli Hamidiye Etfal Hospital between January 2016 and January 2017. Data from a total of 44 patients were retrospectively analyzed.
Device and Technical Parameters
Imaging was performed on two 1.5-T scanners (Avanto, Siemens Healthineers, Erlangen, Germany). SWI was acquired with the following parameters: TR/TE of 49/40 ms, flip angle of 15°, NEX of 1, receiver bandwidth of 80 kHz, matrix size of 256×142, slice thickness of 4 mm, and an interslice gap of 0.8 mm. FLAIR was obtained with the following parameters: TR/TE/TI of 6000/107/2700 ms, flip angle 90°, NSA 3, receiver bandwidth 96 kHz, matrix 256×256, and a 5-mm slice thickness with no interslice gap.
Radiological Evaluation
The CT and MRI images of patients with acute MCA, taken at admission and post-treatment, were evaluated by consensus by two radiologists. On the initial MRI, susceptibility vessel sign (SVS) in the SWI sequence, hyperintense vessel sign (HVS) in FLAIR, and infarct area during hospitalization, as determined by DWI-ASPECTS. The final infarct area was calculated using control CT or FLAIR MRI images taken within the first 7 days after admission. Five patients were excluded because their follow-up MRI was performed beyond one week. Two patients were excluded due to motion artifacts in their imaging, and an additional two patients were excluded because the thrombus could not be localized. Thus, a total of nine patients were excluded from the study.
The NIHSS scores at admission and discharge, the treatment approach, and any additional procedures performed were recorded.
NIHSS (National Institutes of Health Stroke Scale)
NIHSS is an 11-item test, each of which scores a specific skill. While 0 indicates a normal condition for each item, the degree of disability increases as the score increases. All scores are added to reach a total score [10].
ASPECT Score
The use of scoring systems for early ischemic changes in CT increases the success of detecting early ischemic changes. ‘The Alberta Stroke Program Early CT Score (ASPECTS)’ is a system based on CT in evaluation and is a more accurate and practical method compared to the 1/3 MCA rule in ischemic stroke[11]. It is defined on two axial CT slices, one of which is at the thalamus and basal ganglia level, and the second is the section level adjacent to the highest level of ganglionic structures (the first one where they are not visible). Within these two slices, the MCA irrigation area is divided into 10 parts, and each is given one point. Cortical areas are divided into six regions (M1-M6). M1 represents the anterior MCA cortex, M2 represents the MCA cortex lateral to the insular strip, M3 represents the posterior MCA cortex, and M4, M5, and M6 represent the anterior, lateral, and posterior MCA cortex 2 cm above M1, M2, and M3, respectively. In addition, one point is given to the caudate head, lentiform nucleus, internal capsule, and insular strip, giving a total of 10 points. Each region is subtracted one point separately for early ischemic changes such as focal swelling and parenchymal hypodensity. A normal CT examination scores 10 points from the ASPECT score, while a score of 0 indicates widespread involvement of the entire MCA irrigation area.
Statistical Analysis
Statistical analysis was performed using SPSS 15.0 for Windows. Descriptive statistics were presented as numbers and percentages for categorical variables, and as mean, standard deviation, minimum, and maximum for numerical variables. Since numerical variables did not meet the normality assumption, comparisons between two groups were made using the Mann-Whitney U test, and comparisons among more than two groups were made using the Kruskal-Wallis test. The statistical significance level was set at p < 0.05.
Human Rights: Informed consent was obtained.
Ethical Approval
This retrospective study was approved by the Ethics Committee of Sisli Hamidiye Etfal Hospital (Date: 2017-05-16, No: 798). The requirement for informed consent was waived due to the retrospective nature of the study.

Results

The study included 44 patients (24 male, 20 female) with acute (MCA) infarction, averaging 69 years of age. The infarcts were equally distributed between the right and left acute (MCA) infarction territories. Thrombus localization, identified via TOF- MRA, showed the following distribution: MCA M1 segment (20 patients), MCA M2 segment (11 patients), MCA M4 segment (1 patient), and ICA/MCA junction (12 patients).
Key imaging findings revealed that the SVS was detected in 52% of patients (23/44). Thrombus lengths were categorized as 0–10 mm (26 patients) or >10 mm (18 patients). The FLAIR sequence demonstrated hyperintense vessel signs (HVS) in 23 patients. Notably, 87% of patients with SWI SVS also exhibited FLAIR HVS (20/23), indicating a strong positive correlation (r = 0.727) between these markers. Clinically, 30 patients received thrombolysis (tPA), while 14 were managed conservatively. Additional interventions included craniectomy (4 patients) and mechanical thrombectomy (4 patients). Post-treatment hemorrhage occurred in 12 patients.
Outcome analysis highlighted that NIHSS score improvement was most pronounced in the MCA M2 group (mean ΔNIHSS: 3.09) compared to MCA M1 and ICA/MCA groups (ΔNIHSS: 1.5 each), with statistically significant differences (p = 0.005 for M2 vs. M1; p = 0.007 for M2 vs. ICA/MCA). In the patient group with thrombus located in ICA/MCA and MCA M1, the decrease in NIHSS score at the time of discharge was less compared to the group with thrombus located in MCA M2 (Table 1).
Final infarct volumes, assessed via ASPECT scores, were largest in the MCA M2 group (6.64 ± 1.96), followed by MCA M1 (4.50 ± 2.93) and ICA/MCA (3.17 ± 2.17), with M2 vs. ICA/MCA reaching statistical significance (p = 0.002) (Table 2). Thrombus size (0–10 mm vs. >10 mm) did not significantly influence infarct expansion.
Patients exhibiting both SWI SVS and FLAIR HVS demonstrated greater NIHSS improvement (ΔNIHSS: 3.2 vs. 0.88; p = 0.026) compared to those with only one or neither marker. FLAIR HVS alone also predicted better neurological recovery (p = 0.043). However, neither SWI SVS nor FLAIR HVS alone significantly correlated with infarct volume changes (Table 3).
In the patient group in which SWI SVS and FLAIR HVS were detected together, the difference value of ASPECT score detected at the time of discharge and admission was smaller compared to the group in which at most one of these findings was detected, and no significant statistical difference was found between the two groups. On the other hand, in the patient group in which SWI SVS and FLAIR HVS were detected simultaneously, the difference value between the admission and discharge NIHSS scores was larger compared to the group in which at most one of these findings was detected. In the group in which both findings were detected, the difference between the admission and discharge NIHSS scores was significantly larger, and the difference between the two groups was statistically significant (p = 0.026).

Discussion

This study investigated the prognostic value of combined SWI and FLAIR sequences in acute MCA infarction, revealing three key insights. First, the co-presence of SWI susceptibility vessel sign (SVS) and FLAIR hyperintense vessel signs (HVS) was associated with significantly better NIHSS score improvement (ΔNIHSS: 3.20 vs. 0.88; p = 0.026), suggesting these markers may identify patients with robust collateral circulation and tissue perfusion. This aligns with Lee et al. (2009), who proposed FLAIR HVS as a surrogate for leptomeningeal collaterals. Our study extends this by demonstrating synergistic prognostic power when combined with SWI SVS [12].
Second, thrombus location emerged as a critical determinant of outcomes. MCA M2 occlusions had larger infarct volumes (ASPECT: 6.64 ± 1.96) but better clinical recovery (NIHSS: 6.27 ± 6.40) compared to ICA/MCA occlusions (ASPECT: 3.17 ± 2.17; NIHSS: 15.08 ± 6.97; p = 0.007). This paradox may reflect differences in collateral pathways, as MCA M2 segments have richer anastomoses than proximal ICA/MCA junctions [13]. Our findings support the “collateral hypothesis” proposed by Campbell et al. (2019), where distal occlusions permit more effective compensatory flow [8].
Third, while SWI SVS showed moderate sensitivity for thrombus detection (52%), its prognostic utility was significantly enhanced when paired with FLAIR HVS. This mirrors recent work by Koga et al., who found SWI-FLAIR mismatch predicted salvageable penumbra in early-window strokes [4, 14]. However, our small sample size limited statistical power for isolated marker analyses—a limitation also noted in Vural et al. [15].
Notably, FLAIR HVS independently correlated with NIHSS score improvement (p = 0.043), reinforcing its role as a biomarker for viable tissue. Wan et al. similarly linked FLAIR HVS to hyperperfusion after stenting, suggesting it reflects hemodynamic stress in hypoperfused territories [16]. Our data extend this concept to natural history stroke, where HVS may flag regions with perfusion reserves.
Technical considerations merit discussion. SWI’s sensitivity to deoxyhemoglobin allowed thrombus detection even in small vessels (e.g., MCA M4), but its prognostic value for infarct expansion was limited. This contrasts with a study that associated susceptibility vessel signs with larger final infarcts [17]. Our null finding may reflect cohort differences or the use of ASPECT scores (vs. volumetric analysis).
Finally, the lack of perfusion imaging in our protocol is a limitation. Recent studies (e.g., DEFUSE-3 [18]) emphasize perfusion-diffusion mismatch for patient selection. Future work should integrate SWI-FLAIR findings with perfusion maps to refine predictive models.
The relatively small sample size (n = 44) may have limited the statistical power to detect subtle associations, particularly in subgroup analyses. This constraint is common in single-center retrospective studies but could affect the generalizability of our findings. Infarct volume assessment using ASPECT scores, while clinically practical, lacks the precision of volumetric measurements. Quantitative volumetric analysis might have revealed more nuanced relationships between imaging markers and tissue outcomes. The absence of perfusion imaging data represents a significant gap, as perfusion-diffusion mismatch is now recognized as critical for understanding penumbral salvageability. This prevented direct correlation between our SWI-FLAIR findings and tissue perfusion status. The retrospective design introduced potential selection biases, particularly in treatment allocation (e.g., only 30 patients received tPA). Prospective randomization would strengthen causal inferences.

Limitations

Multicenter studies with larger cohorts (n > 200) are needed to validate our findings regarding the synergistic prognostic value of combined SWI SVS and FLAIR HVS markers. The STROBE guidelines should be followed to ensure methodological rigor. Longitudinal studies with 90-day mRS outcomes would better characterize the long-term clinical relevance of these imaging biomarkers beyond discharge NIHSS scores. Artificial intelligence approaches could automate the detection of SWI SVS and FLAIR HVS patterns, potentially developing predictive models for clinical recovery using deep learning.

Conclusion

This study demonstrates that combined SWI SVS and FLAIR HVS detection predicts better clinical outcomes in acute MCA infarction, likely by identifying preserved collateral flow and tissue perfusion.
Future research should focus on several key areas. First, standardization of SWI acquisition protocols and interpretation criteria is essential to optimize both sensitivity and specificity. Second, artificial intelligence approaches could automate the detection of SWI SVS and FLAIR HVS patterns, potentially developing predictive models for clinical recovery. Third, integration with perfusion imaging would provide comprehensive assessment of tissue viability and collateral status. Finally, prospective validation in larger multicenter cohorts is needed to establish the clinical utility of this multimodal approach.

References

1. Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the early management of patients with acute ischemic stroke: 2019 update to the 2018 guidelines. Stroke. 2019;50(12):e344–e418
   PubMed Google Scholar
2. Virani SS, Alonso A, Aparicio HJ, et al. Heart disease and stroke statistics–2021 update: a report from the American Heart Association. Circulation. 2021;143(8):e254–e743.
   PubMed Google Scholar
3. Hasan TF, Hasan H, Kelley RE. Overview of acute ischemic stroke evaluation and management. Biomedicines. 2021;9(10):1486.
   PubMed Google Scholar
4. Kim BJ, Han MK, Park TH, et al. Magnetic resonance imaging in acute ischemic stroke treatment. J Stroke. 2014;16(3):131–45.
   PubMed Google Scholar
5. Wardlaw JM, Murray V, Berge E, et al. Thrombolysis for acute ischaemic stroke. Cochrane Database Syst Rev. 2014;(7):1-143.
   PubMed Google Scholar
6. Lansberg MG, Straka M, Kemp S, et al. RAPID automated patient selection for reperfusion therapy: a pooled analysis. Stroke. 2011;42(6):1608–14.
   PubMed Google Scholar
7. Hasan TF, Rabinstein AA, Gronseth GS, et al. Diagnosis and management of acute ischemic stroke. Mayo Clin Proc. 2018;93(4):523–38.
   PubMed Google Scholar
8. Campbell BCV, Majoie CBLM, Albers GW, et al. Extending thrombolysis to 4.5–9 h and wake-up stroke using perfusion imaging. Lancet. 2019;394(10193):139–47.
   PubMed Google Scholar
9. Lummel N, Schmitt WR, Dechent P, et al. 3D fluid-attenuated inversion recovery imaging: reduced CSF artifacts and enhanced detection of subarachnoid hemorrhage. AJNR Am J Neuroradiol. 2011;32(11):2054–60.
   PubMed Google Scholar
10. Tong DC, Yenari MA, Albers GW, et al. Correlation of perfusion- and diffusion- weighted MRI with NIHSS score in acute (< 6.5-hour) ischemic stroke. Neurology. 1998;50(4):864–70.
    PubMed Google Scholar
11. Pexman JH, Barber PA, Hill MD, et al. Use of the Alberta Stroke Program Early CT Score (ASPECTS) for assessing CT scans in acute stroke. AJNR Am J Neuroradiol. 2001;22(8):1534–42.
    PubMed Google Scholar
12. Lee KY, Latour LL, Luby M, et al. Distal hyperintense vessels on FLAIR: an MRI marker for collateral circulation in acute stroke? Neurology. 2009;72(13):1134–9.
    PubMed Google Scholar
13. Amukotuwa SA, Straka M, Mlynash M, et al. Distal medium vessel occlusions can be accurately and rapidly detected using Tmax maps. Stroke. 2021;52(10):3308–17.
    PubMed Google Scholar
14. Koga M, Kimura K, Shibazaki K, et al. Impact of stroke severity and vascular risk factors on early versus late anticoagulation in patients with stroke and atrial fibrillation. J Stroke. 2025;27(2):284–8.
    PubMed Google Scholar
15. Vural A, Zileli R, Inan N, et al. Bright and dark vessels on stroke imaging: different sides of the same coin? Diagn Interv Radiol. 2016;22(3):284–90.
    PubMed Google Scholar
16. Wan CC, Yeh LR, Chen CC, et al. FLAIR vascular hyperintensities predict cerebral hyperperfusion after intracranial stenting. Neuroradiology. 2017;59(8):791–6.
    PubMed Google Scholar
17. Kranz PG, Tanpitukpongse TP, Choudhury KR, et al. Imaging signs in spontaneous intracranial hypotension: prevalence and relationship to CSF pressure. AJNR Am J Neuroradiol. 2016;37(7):1374–8.
    PubMed Google Scholar
18. Sarraj A, Hassan AE, Grotta JC, et al. Outcomes of thrombectomy in transferred patients with ischemic stroke in the late window: a subanalysis from the DEFUSE 3 trial. JAMA Neurol. 2019;76(6):682–9.
    PubMed Google Scholar

Declarations

Additional Information

Publisher’s Note
Bayrakol MP remains neutral with regard to jurisdictional and institutional claims.

Rights and Permissions

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). To view a copy of the license, visit https://creativecommons.org/licenses/by-nc/4.0/

View full license details →

About This Article