Retinal sub-layer thicknesses, presence of lacunes, and their interaction with cognitive performance in recent single subcortical infarction

To the Editor: Recent single subcortical infarction (RSSI), formerly known as lacunar stroke, is a neurological syndrome that occurs when a small perforating artery in the brain becomes blocked, resulting in ischemia and neurological impairment.[1] Radiological indicators of cerebral small vessel disease (SVD), such as lacunes and white matter hyperintensity (WMH), have gained increasing attention because of their role in cognitive impairment and dementia.[1] The correlation among radiological indicators enables the investigation of their impact on both structural and functional damage to the brain. The brain and the retina share many characteristics such as embryologic origin, precise neuronal cell layers, and microvasculature.[2,3] The presence of ocular manifestations in RSSI and other forms of ischemic stroke emphasizes the strong relationship between the retina and the brain.[2,3] Quantitative changes in the retinal structure such as thinning of the retinal layers have been associated with RSSI and cerebral pathologies.[2,3] However, data on the association between both retinal structural changes and radiological markers in RSSI and how these factors jointly influence cognition are lacking. This study aimed to investigate the association between optical coherence tomography (OCT) metrics and the presence of lacunes on cognitive performance in RSSI patients. The study was approved by The West China Hospital of Sichuan University Ethics Committee (No. 2020[922]). All participants provided written informed consent before enrolling in the study. A total of 132 RSSI patients between January 2021 and October 2023 underwent magnetic resonance imaging, and SVD markers were assessed [Figure 1A]. SVD MRI markers such as lacunes, WMH, and enlarged perivascular spaces (PVS) were evaluated according to the Standards for ReportIng Vascular changes on Neuroimaging (STRIVE) consensus criteria.[1] The swept-source (SS)-OCT (VG200S; SVision Imaging, Henan, China; version 2.0.106) was used for retinal imaging in all participants. The specifications of the tool have been well described in our previous report.[4] For retinal structural thicknesses, automatic segmentation of the retinal thickness was done by the OCT tool. Here, we analyzed the retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GCIPL) in a 3 mm × 3 mm[2] area around the fovea in the macula as shown in Figure 1A. The RNFL was defined as the thickness between the base of the inner limiting membrane (ILM) to the top border of the ganglion cell layer (GCL). GCIPL was defined as the thickness from the base of the RNFL to the top border of the inner nuclear layer (INL). The Beijing version of the Montreal Cognitive Assessment (MoCA-BJ) was performed on all participants by a well-trained physician. In our study, cutoff MoCA-BJ score was 22 for the detection of cognitive impairment. Participants with MoCA scores ≥22 were characterized as cognitively normal (non-cognitively impaired [NCI]), while those with MoCA scores ≤21 were characterized as cognitively impaired (CI). Of the 132 RSSI patients included in our study, 81 were NCI while 51 were CI. Demographic and clinical characteristics and data analysis are shown in the Supplementary Tables 1−3, https://links.lww.com/CM9/C82. CI patients had thinner RNFL (Z = −2.290, P = 0.022) thickness compared to NCI as shown in the Supplementary Table 1, https://links.lww.com/CM9/C82; no significant difference was seen in the GCIPL thickness when both groups were compared (Z = −1.011, P = 0.313). RNFL (β = 0.366, P = 0.042) and GCIPL (β = 0.093, P = 0.034) thicknesses showed significant correlations with MoCA scores as shown in Figure 1B and Supplementary Table 2, https://links.lww.com/CM9/C82. Similarly, the presence of lacunes (β = −1.478, P = 0.029) significantly correlated with MoCA scores. In this RSSI cohort, there was a significant interaction between RNFL (β = 0.712, P = 0.036) and GCIPL thicknesses (β = 0.209, P = 0.013) and the presence of lacunes on MoCA scores, respectively [Supplementary Table 3, https://links.lww.com/CM9/C82].Figure 1: (A) Representative image of MRI markers and OCT metrics. RSSI in the basal ganglia on DWI (a). Moderate to severe PVS in the basal ganglia on T2-weighted imaging (b). High-grade periventricular WMH (Fazekas score 3) (c) and deep WMH (Fazekas score 2) (d) on FLAIR. Two chronic lacunes in the right centrum semiovale (arrowheads) on FLAIR (d) and T1-weighted imaging (e). The RNFL was defined as the thickness between the base of the ILM to the top border of the GCL; RNFL is shown as the thickness between the two red lines. GCIPL was defined as the thickness from the base of the RNFL to the top border of the INL. GCIPL is represented as the thickness between the lower red line and the blue line (f). (B) Correlation between MoCA scores and OCT metrics and presence of lacunes in RSSI patients. A positive correlation was seen between MoCA scores and RNFL (a) and GCIPL (b) thicknesses respectively. A negative correlation was seen between MoCA scores and the presence of lacunes (c). The X-axis and Y-axis of added variable plots were the residuals of the dependent variable and the independent variable when both of these variables were regressed on risk factors. DWI: Diffusion-weighted imaging; FLAIR: Fluid-attenuated inversion recovery; GCIPL: Ganglion cell-inner plexiform layer; GCL: Ganglion cell layer; ILM: Inner limiting membrane; INL: Inner nuclear layer; MoCA: Montreal Cognitive Assessment; MRI: Magnetic resonance imaging; OCT; Optical coherence tomography; PVS: Perivascular space; RNFL: Retinal nerve fiber layer; RSSI: Recent single subcortical infarction; WMH: White matter hyperintensity.Changes in the retinal structural thicknesses are suggested to reflect related neurodegeneration occurring in the brain.[3,4] Retinal imaging studies have shown that cerebral infarction patients have thinner RNFL thickness due to neurodegeneration.[3] This study showed that patients with CI have thinner RNFL thickness compared to NCI. The retinal sub-layer thicknesses are measures of retinal ganglion cell integrity; RNFL includes the retinal ganglion cell axons whilst the GCIPL contains the cell bodies and dendrites of the retinal ganglion cells.[4] Thinning of these retinal structural layers reflects neurodegeneration and is suggested to be associated with cognitive dysfunction. There is increasing evidence that cognitive impairment occurs after RSSI through microvascular dysfunction.[5] Given that microvascular dysfunction is associated with neurodegeneration and the retinal microvasculature is responsible for the metabolism of retinal neuronal integrity, RNFL thinning in CI compared to NCI may suggest that retinal neurodegeneration may be more severe in RSSI patients with CI. This study found a substantial correlation between the MoCA scores of RSSI patients and the GCIPL and RNFL, suggesting that OCT measurement may be a useful tool for identifying cognitive deterioration in these individuals. This study also showed that the presence of lacunes was significantly associated with cognitive function in RSSI patients. The cognitive function in RSSI associated with the presence of lacunes may well be due to damage to cortical-subcortical pathways, disrupting the complex and distributed networks that underpin cognition. The study showed an interaction between RNFL and GCIPL thicknesses and the presence of lacunes on MoCA scores in RSSI patients, respectively. In the retina, thinning of the RNFL and GCIPL (neurodegeneration) is suggested to be associated with reduced blood flow as a result of ischemia.[2,5] A similar mechanism involving hypoxia and/or cerebral ischemia has been proposed to contribute to the formation of lacunes. These similarities suggest that comparable mechanisms may occur concurrently in both the retina and brain of RSSI patients. Thus, the interaction between RNFL and GCIPL thicknesses and the presence of lacunes on cognitive performance in RSSI patients suggests that changes in OCT metrics (retinal structural thicknesses) and the presence of lacunes may jointly influence cognitive performance. Our findings have important clinical implications. Cognitive impairment after stroke is one of the major determinants of functional dependence in stroke survivors. As the pathophysiology and trajectory of cognitive decline after stroke are complex, with numerous determinants, thus, there is a need for biomarkers that may be sensitive to cognitive changes. An important finding in our study was the association between OCT metrics and the presence of lacunes with cognitive performance in RSSI patients. These findings suggest that changes in the retina and brain may occur concurrently leading to cognitive changes in RSSI patients. Therefore, retinal changes after RSSI in clinics should not be neglected. The retina and brain should be considered as means to observe pathophysiological changes associated with cognitive changes after RSSI. These findings may enable the assessment of treatment at an earlier stage. In conclusion, our study with quantitative measures of the retinal structural thicknesses and presence of lacunes in RSSI patients provides insight into temporal dynamics of neurodegeneration in the retina and brain on cognitive function. The clinical utility of OCT metrics and the presence of lacunes in predicting cognitive impairment in RSSI patients may require further study. Funding This work was supported by grants from the National Natural Science Foundation of China (Nos. 82271328, 82071320, 82371322, and 82301661), China Postdoctoral Science Foundation (Nos. 2022M712249 and 2023T160447), Post-Doctor Research Project, West China Hospital, Sichuan University (No. 2023HXBH007), National Key R&D Program of China (Nos. 2023YFC2506600, 2023YFC2506603), and the 1·3·5 Project for Disciplines of Excellence–Clinical Research Incubation Project of West China Hospital at Sichuan University (No. 2020HXFH012). Conflicts of interest None.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.