Yang Tang , Huajiao Wang , Xiaohua Wang , Zichun Yan , Qiyuan Zhu , Ting Yin , Yuhui Xu , Yiqiu Wei , Bin Yang , Zhuowei Shi , Yongmei Li
{"title":"多发性硬化症FLAIR超边缘病变与顺磁边缘病变的关系","authors":"Yang Tang , Huajiao Wang , Xiaohua Wang , Zichun Yan , Qiyuan Zhu , Ting Yin , Yuhui Xu , Yiqiu Wei , Bin Yang , Zhuowei Shi , Yongmei Li","doi":"10.1016/j.jneuroim.2025.578738","DOIUrl":null,"url":null,"abstract":"<div><div>Paramagnetic rim lesions (PRLs), identified using susceptibility-sensitive sequences, are established prognostic imaging markers for multiple sclerosis (MS). However, susceptibility-sensitive sequences are not yet routinely performed in many clinical centers. We aim to investigate an imaging feature observed on conventional T2 fluid-attenuated inversion recovery (FLAIR) sequences, termed “FLAIR hyper-rim”, and explore its association with PRLs. This study included 61 relapsing-remitting MS (RRMS) and 35 healthy controls. Based on the presence or absence of paramagnetic rim in susceptibility-sensitive images (PRL+/−) and FLAIR hyper-rim sign in T2-FLAIR images (Rim+/−), white matter (WM) lesions were classified into four subgroups: (1) PRL + Rim+; (2) PRL + Rim-; (3) PRL-Rim+; (4) PRL-Rim-. Differences in lesion volume and microstructural damage were compared, based on diffusion kurtosis imaging (DKI) parameters, including kurtosis fractional anisotropy (KFA), mean kurtosis (MK), axial kurtosis (AK), and radial kurtosis (RK). The correlations between lesion load and clinical scale scores and brain volume were further investigated. 1109 WM lesions were analyzed, including 338 PRLs and 300 FLAIR hyper-rim lesions. Of 300 FLAIR hyper-rim lesions, 197 (65.7 %) co-localized with a PRL, and 197/338 (58.3 %) PRLs co-localized with a FLAIR hyper-rim lesion. Considering the chance-level overlap, we further calculated the Jaccard index (44.7 %, 95 % CI: 39.8–49.5), and Cohen's kappa (κ = 0.463, 95 % CI: 0.408–0.518), both indicating a moderate association. The PRL + Rim+ group exhibited the lowest AK and RK values, significantly different from those of other three groups (all <em>P</em> < 0.05). Additionally, the PRL-Rim+ group exhibited lower MK, AK and RK values compared to the PRL-Rim- group (all <em>P</em> < 0.05). Furthermore, FLAIR hyper-rim lesions were negatively related to cognitive test and brain volume. FLAIR hyper-rim lesions appear to represent a subtype of MS lesions characterized by more severe tissue damage and may help identify lesions more likely to be PRLs, although their predictive value requires further validation.</div></div>","PeriodicalId":16671,"journal":{"name":"Journal of neuroimmunology","volume":"408 ","pages":"Article 578738"},"PeriodicalIF":2.5000,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Characterizing the relationship between FLAIR hyper-rim lesions and paramagnetic rim lesions in multiple sclerosis\",\"authors\":\"Yang Tang , Huajiao Wang , Xiaohua Wang , Zichun Yan , Qiyuan Zhu , Ting Yin , Yuhui Xu , Yiqiu Wei , Bin Yang , Zhuowei Shi , Yongmei Li\",\"doi\":\"10.1016/j.jneuroim.2025.578738\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Paramagnetic rim lesions (PRLs), identified using susceptibility-sensitive sequences, are established prognostic imaging markers for multiple sclerosis (MS). However, susceptibility-sensitive sequences are not yet routinely performed in many clinical centers. We aim to investigate an imaging feature observed on conventional T2 fluid-attenuated inversion recovery (FLAIR) sequences, termed “FLAIR hyper-rim”, and explore its association with PRLs. This study included 61 relapsing-remitting MS (RRMS) and 35 healthy controls. Based on the presence or absence of paramagnetic rim in susceptibility-sensitive images (PRL+/−) and FLAIR hyper-rim sign in T2-FLAIR images (Rim+/−), white matter (WM) lesions were classified into four subgroups: (1) PRL + Rim+; (2) PRL + Rim-; (3) PRL-Rim+; (4) PRL-Rim-. Differences in lesion volume and microstructural damage were compared, based on diffusion kurtosis imaging (DKI) parameters, including kurtosis fractional anisotropy (KFA), mean kurtosis (MK), axial kurtosis (AK), and radial kurtosis (RK). The correlations between lesion load and clinical scale scores and brain volume were further investigated. 1109 WM lesions were analyzed, including 338 PRLs and 300 FLAIR hyper-rim lesions. Of 300 FLAIR hyper-rim lesions, 197 (65.7 %) co-localized with a PRL, and 197/338 (58.3 %) PRLs co-localized with a FLAIR hyper-rim lesion. Considering the chance-level overlap, we further calculated the Jaccard index (44.7 %, 95 % CI: 39.8–49.5), and Cohen's kappa (κ = 0.463, 95 % CI: 0.408–0.518), both indicating a moderate association. The PRL + Rim+ group exhibited the lowest AK and RK values, significantly different from those of other three groups (all <em>P</em> < 0.05). Additionally, the PRL-Rim+ group exhibited lower MK, AK and RK values compared to the PRL-Rim- group (all <em>P</em> < 0.05). Furthermore, FLAIR hyper-rim lesions were negatively related to cognitive test and brain volume. FLAIR hyper-rim lesions appear to represent a subtype of MS lesions characterized by more severe tissue damage and may help identify lesions more likely to be PRLs, although their predictive value requires further validation.</div></div>\",\"PeriodicalId\":16671,\"journal\":{\"name\":\"Journal of neuroimmunology\",\"volume\":\"408 \",\"pages\":\"Article 578738\"},\"PeriodicalIF\":2.5000,\"publicationDate\":\"2025-08-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of neuroimmunology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S016557282500219X\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"IMMUNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of neuroimmunology","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S016557282500219X","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"IMMUNOLOGY","Score":null,"Total":0}
Characterizing the relationship between FLAIR hyper-rim lesions and paramagnetic rim lesions in multiple sclerosis
Paramagnetic rim lesions (PRLs), identified using susceptibility-sensitive sequences, are established prognostic imaging markers for multiple sclerosis (MS). However, susceptibility-sensitive sequences are not yet routinely performed in many clinical centers. We aim to investigate an imaging feature observed on conventional T2 fluid-attenuated inversion recovery (FLAIR) sequences, termed “FLAIR hyper-rim”, and explore its association with PRLs. This study included 61 relapsing-remitting MS (RRMS) and 35 healthy controls. Based on the presence or absence of paramagnetic rim in susceptibility-sensitive images (PRL+/−) and FLAIR hyper-rim sign in T2-FLAIR images (Rim+/−), white matter (WM) lesions were classified into four subgroups: (1) PRL + Rim+; (2) PRL + Rim-; (3) PRL-Rim+; (4) PRL-Rim-. Differences in lesion volume and microstructural damage were compared, based on diffusion kurtosis imaging (DKI) parameters, including kurtosis fractional anisotropy (KFA), mean kurtosis (MK), axial kurtosis (AK), and radial kurtosis (RK). The correlations between lesion load and clinical scale scores and brain volume were further investigated. 1109 WM lesions were analyzed, including 338 PRLs and 300 FLAIR hyper-rim lesions. Of 300 FLAIR hyper-rim lesions, 197 (65.7 %) co-localized with a PRL, and 197/338 (58.3 %) PRLs co-localized with a FLAIR hyper-rim lesion. Considering the chance-level overlap, we further calculated the Jaccard index (44.7 %, 95 % CI: 39.8–49.5), and Cohen's kappa (κ = 0.463, 95 % CI: 0.408–0.518), both indicating a moderate association. The PRL + Rim+ group exhibited the lowest AK and RK values, significantly different from those of other three groups (all P < 0.05). Additionally, the PRL-Rim+ group exhibited lower MK, AK and RK values compared to the PRL-Rim- group (all P < 0.05). Furthermore, FLAIR hyper-rim lesions were negatively related to cognitive test and brain volume. FLAIR hyper-rim lesions appear to represent a subtype of MS lesions characterized by more severe tissue damage and may help identify lesions more likely to be PRLs, although their predictive value requires further validation.
期刊介绍:
The Journal of Neuroimmunology affords a forum for the publication of works applying immunologic methodology to the furtherance of the neurological sciences. Studies on all branches of the neurosciences, particularly fundamental and applied neurobiology, neurology, neuropathology, neurochemistry, neurovirology, neuroendocrinology, neuromuscular research, neuropharmacology and psychology, which involve either immunologic methodology (e.g. immunocytochemistry) or fundamental immunology (e.g. antibody and lymphocyte assays), are considered for publication.