To the Editor: Information processing speed (IPS), typically assessed using the Symbol Digit Modalities Test (SDMT), is a highly sensitive yet fragile cognitive domain that often mediates age-related decline in other cognitive abilities. 1, 2 White matter hyperintensities (WMHs) are a prominent feature of cerebral small vessel disease and have been reported to mediate the association between blood–brain barrier dysfunction and IPS decline in older adults. 3 Given that IPS performance relies heavily on attentional control and executive processing, disruptions within attention- and executive-related brain networks may play a critical role in WMH-related decline in IPS. Although cognitive performance tends to deteriorate as WMH burden increases, there is often a notable discrepancy between the severity of WMH observed on neuroimaging and the actual degree of cognitive impairment. Some individuals maintain relatively preserved IPS despite substantial WMH lesions, which may be due to population heterogeneity, variations in lesion distribution, or confounding comorbid clinical factors. These observations suggest that WMH-related cognitive decline involves complex compensatory or adaptive mechanisms. However, such mechanisms are currently poorly understood. Therefore, we hypothesized that individuals with WMHs who exhibit varying degrees of IPS would show distinct patterns of functional connectivity (FC) and differences in network-level information transfer efficiency, which may provide insight into the dissociation between WMH burden and cognitive outcomes. To test this, we conducted a case-control neuroimaging study investigating the functional network correlates of IPS variability in individuals with WMHs using resting-state functional magnetic resonance imaging (MRI) and independent component analysis (ICA). We enrolled 145 right-handed participants diagnosed with WMHs at Yueyang Hospital of Integrated Traditional Chinese and Western Medicine between April 2020 and April 2023. Based on a 1 standard deviation cutoff of the mean SDMT score, participants were classified into a low-SDMT score (SLS; ≤28, n = 65) or high-SDMT score (SHS; >28, n = 80) group. The study was approved by the Medical Ethics Committee of the Yueyang Hospital of Integrated Traditional Chinese and Western Medicine, Shanghai University of Traditional Chinese Medicine (No. 2019-124), registered in the Chinese Clinical Trial Registry (www. chictr. org. cn, ChiCTR2000031744), and conducted with written informed consent from all participants. The main inclusion criteria were as follows: (1) total Fazekas scale score ≥1; (2) aged between 50 and 75 years; (3) mini-mental state examination (MMSE) score ≥24; (4) Hamilton Depression Rating Scale score 0. 05). Significant between-group differences were observed in several neuropsychological domains. Compared with the SLS group, the SHS group showed better memory performance, as reflected by higher scores on the Auditory Verbal Learning Test (AVLT, Z = 2. 775, P = 0. 006) and its delayed recall component (AVLTN5, Z = 2. 510, P = 0. 012). IPS was also superior in the SHS group compared with the SLS group, as indicated by higher SDMT scores (Z = 10. 340, P <0. 001) and shorter completion times on the Shape Trails Test Part B (Z = –5. 740, P <0. 001). In addition, the SHS group outperformed the SLS group in the language assessments, as measured by the Boston Naming Test (BNT, Z = 3. 325, P = 0. 001) Supplementary Table 1, https: //links. lww. com/CM9/C828. Correlation analyses revealed that Fazekas scores were negatively associated with SDMT performance and positively correlated with age (all P <0. 05, FDR-corrected), whereas no other cognitive measures (i. e. , MMSE, AVLT, Shape Trails Test Part B (STT-B), BNT, Hamilton Depression Rating Scale, or HAMA scores) showed significant correlations Supplementary Figure 3, https: //links. lww. com/CM9/C828. A total of nine functional networks were identified using ICA, and the network atlas is presented in Supplementary Figure 4, https: //links. lww. com/CM9/C828. Correlation analyses revealed significant associations between WMH severity and network-level properties. Specifically, the results revealed that the global efficiency (Eglob) of the ventral attention network (vAN) was negatively correlated with the Fazekas scores (r = –0. 161, P <0. 001, FDR-corrected), whereas the characteristic path length of the vAN showed a positive correlation (r = 0. 214, P <0. 001, FDR-corrected). No other networks were significantly correlated with global properties or Fazekas scores Supplementary Table 2 and Supplementary Figure 5, https: //links. lww. com/CM9/C828. In addition, the FC between the right frontoparietal network (rFPN) and dorsal attention network (dAN) was negatively correlated with Fazekas scores (r = –0. 157, P = 0. 003, FDR-corrected). The corresponding scatterplots are presented in Supplementary Figure 6, https: //links. lww. com/CM9/C828. Participants with higher SDMT scores exhibited significantly stronger FC within the vAN than those with lower SDMT scores. Specifically, the higher FC was observed in the right lingual gyrus (P <0. 05, FDR-corrected). Moreover, FC in this region was positively correlated with SDMT performance across all participants (P <0. 05). Inter-network analyses further revealed increased FC in the SHS group between the rFPN and both the medial visual network (mVN) and dAN (P <0. 05, FDR-corrected) Figure 1. No intra-network FC differences survived correction for multiple comparisons Supplementary Figure 7 and Supplementary Table 3, https: //links. lww. com/CM9/C828. Figure 1: Key intra- and inter-network FC differences in IPS. (A) Brain regions showing reduced intra-network FC in the SLS compared with the SHS group. (B) Correlation between regional FC and SDMT scores. (C) Heat map of inter-network FC averaged across subjects. (D) Brain renderings illustrating significant inter-network FC differences between groups. (E) Group comparisons of inter-network FC between SLS and SHS. * P <0. 05. AN: Auditory network; DMN: Default mode network; FC: Functional connectivity; FPN: Frontoparietal network; IPS: Information processing speed; SDMT: Symbol Digit Modalities Test; SHS: High SDMT score; SLS: Low SDMT score; SMN: Sensorimotor network; VN: Visual network. Prefixes d, v, p, r, and m indicate dorsal, ventral, posterior, right, and medial, respectively. The area under the curve (AUC) -based analysis of global property parameters yielded no statistically significant between-group differences. However, interaction analyses between group and global properties for cognitive functions revealed significant associations Supplementary Figure 8, https: //links. lww. com/CM9/C828. Specifically, vAN small-worldness AUC exhibited significant effects on both the SDMT and STTB scores (P <0. 01 and P <0. 001, respectively, FDR-corrected). In addition, rFPNₗocal efficiency (Eloc) AUC had a significant interaction effect on SDMT and MMSE scores (P <0. 01 and P <0. 001, respectively, FDR corrected), rFPNclustering coefficient (Ecp) AUC had a significant interaction effect on SDMT and MMSE scores (P <0. 001 and P <0. 001, respectively, FDR-corrected), and rFPNEglobAUC had a significant interaction effect on MMSE scores (P <0. 001, FDR-corrected). More detailed results are available in the Supplementary Matreials, https: //links. lww. com/CM9/C828. This study examined the functional brain network mechanisms underlying differences in IPS in individuals with WMHs. Higher Fazekas scores were associated with lower SDMT performance and older age, indicating that WMH burden contributes to cognitive slowing. 3 Of the nine major resting-state networks, only the vAN (particularly the lingual gyrus) showed reduced intra-network FC in participants with slower IPS. Although not a canonical vAN node, the lingual gyrus connects visual and parieto-frontal regions, suggesting its involvement in the integration of visual and attentional information. The inter-network analyses revealed lower FC between the rFPN and both the dAN and mVN in those with slower IPS. The rFPN is critical for executive control and interacts with the dAN to support top-down attentional modulation, demonstrating that WMHs preferentially impair frontoparietal and attentional systems. Although global network metrics did not differ between IPS groups, generalized linear modeling indicated that Eglob of the vAN and local efficiency of the rFPN influenced IPS performance, which highlights the functional relevance of these networks. These results emphasized the importance of attentional/executive network communication efficiency in determining why some individuals maintain IPS despite exhibiting WMHs. Our findings provide new insights into the neural basis of cognitive slowing in those with WMHs and highlight the critical role of attentional and executive networks in maintaining information-processing efficiency. This study has several limitations: the absence of detailed lesion-location analysis, heterogeneity in WMH severity, and the cross-sectional design. These factors should be considered when interpreting the results because they may affect the accuracy of region-specific inferences. Future studies incorporating lesion mapping and longitudinal data would help validate our findings. In conclusion, we found that in patients with WMHs, decreased FC and reduced communication efficiency within attentional and frontoparietal networks are associated with impaired IPS. These network-level associations may contribute to the discrepancy between white matter lesion burden and cognitive performance. Funding This work was supported by grants from the National Natural Science Foundation of China (Nos. 82302870 and 82472589) ; Shanghai Rising-Star Program (No. 24QA2709300) ; Shanghai University of Traditional Chinese Medicine Science and Technology Development Fund (No. 23KFL112) ; and Shanghai Oriental Talents Program–Youth Project (No. QNJY2024077). Conflicts of interest None.
Jin et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: