Pathophysiology of cerebral small vessel disease: a journey through recent discoveries

Abstract

Cerebral small vessel disease (cSVD) encompasses a heterogeneous group of age-related small vessel pathologies that affect multiple regions. Disease manifestations range from lesions incidentally detected on neuroimaging (white matter hyperintensities, small deep infarcts, microbleeds, or enlarged perivascular spaces) to severe disability and cognitive impairment. cSVD accounts for approximately 25% of ischemic strokes and the vast majority of spontaneous intracerebral hemorrhage and is also the most important vascular contributor to dementia. Despite its high prevalence and potentially long therapeutic window, there are still no mechanism-based treatments. Here, we provide an overview of the recent advances in this field. We summarize recent data highlighting the remarkable continuum between monogenic and multifactorial cSVDs involving NOTCH3, HTRA1, and COL4A1/A2 genes. Taking a vessel-centric view, we discuss possible cause-and-effect relationships between risk factors, structural and functional vessel changes, and disease manifestations, underscoring some major knowledge gaps. Although endothelial dysfunction is rightly considered a central feature of cSVD, the contributions of smooth muscle cells, pericytes, and other perivascular cells warrant continued investigation.

Figure 1. Neuroimaging features of cSVDs.

In 2013, a group of experts published Standards for Reporting Vascular Changes on Neuroimaging (STRIVE-1) (2) — an attempt to harmonize terminology and definitions of key MRI features associated with cSVDs. These features include the following: white matter hyperintensities (WMHs) on T2-weighted MRI sequences (yellow); recent, small subcortical infarcts; subcortical lacunes of presumed vascular origin (3–15 mm fluid-filled cavities) (dark tan), likely the end result of a small subcortical infarct or microhemorrhage; perivascular (fluid-filled) spaces that follow the course of small perforating vessels (purple); microbleeds (2–5 mm diameter), detected as hypointense lesions on T2* images or susceptibility-weighted sequences (red); intracerebral hemorrhage (ICH) (red); and brain atrophy.

Figure 2. Integrated representation of the anatomy, cellular composition, and physiology of brain vessels.

(A) Schematic of the arteriovenous axis with the four main vascular compartments, including the artery/arteriole, the arteriole-capillary transition (ACT) zone, the capillary bed and the venule/vein, and their associated cells: arterial endothelial cells (aECs), arterial SMCs (aSMCs), transitional cells (trans cells, orange), capillary endothelial cells (capECs), venous endothelial cells (vECs), and venous SMCs (vSMCs). Penetrating arteries and arterioles are separated from the brain parenchyma by a fluid-filled space (light green) that disappears as arterioles morph into capillaries and then reappears around veins. The perivascular space (inset) contains resident cells (PVMs and perivascular fibroblasts, PVFBs) and is delimited on the parenchymal side by the glia limitans formed by astrocytic endfeet. (B) Simplified depiction of the main brain vessel functions with respect to each vascular compartment. From top to bottom: (i) CBF autoregulation increases or decreases vessel diameter in response to BP decreases and increases, respectively. aSMCs are the primary sensors of BP changes and the primary effector cells driving changes in vessel diameter. (ii) Neurovascular coupling starts with the increase in local neural activity that leads to capEC hyperpolarization. Hyperpolarizing signal is propagated to upstream arterioles/arteries and transmitted to aSMCs, resulting in retrograde vasodilation. (iii) The BBB is formed by ECs, mural cells with their basement membrane, and astrocytic endfeet. Tight junctions between ECs prevent free paracellular transport of molecules; ECs express specific influx transporters and efflux pumps, which drive the active transport of specific solutes and metabolites into or out of the brain, respectively, and are enriched for the lipid transporter MFSD2A, which inhibits the rate of transcytosis (113, 166). (iv) The glymphatic system involves (a) CSF influx along the periarterial spaces, driven mainly by arterial pulsatility; (b) CSF entry into the brain supported by aquaporin 4 (AQP4) channel expression on the astrocytic endfeet, subsequent mix with the ISF, and flow through the extracellular spaces; and (c) the efflux of extracellular fluid and wastes along perivenous spaces.

Figure 3. Opposite changes in mural cells in the arteriole-capillary transition zone are associated with distinct cSVD features.

Schematic representation of brain vessels in (A) the collagen IV–related cSVD, which manifests as recurrent spontaneous ICHs, and in (B) the NOTCH3 null–driven cSVD, which is characterized by recurrent deep infarcts. (A) In the collagen IV disease, the ACT zone shows an increased number of mural cells with higher contractile protein content, raising intravascular pressure in the upstream feeding arteriole, which exhibits loss of SMCs, and promoting arteriolar rupture at the site of SMC loss and hemorrhage (red). (B) In the NOTCH3 null–driven cSVD, the combination of loss of arterial SMCs and loss of mural cells in the ACT zone is predicted to decrease perfusion pressure and promote ischemic lesions in the deep brain regions (gray).