Magnetic Resonance Imaging Markers of Cerebral Small Vessel Disease in Adults with Moyamoya Disease.

Abstract

In clinical work, the magnetic resonance imaging markers of cerebral small vessel disease (CSVD) are frequently observed in moyamoya disease (MMD), but the clinical significance of these markers in MMD remains unclear. This study aimed to fill this gap and systematically investigate its clinical significance. In this retrospective cohort study, we screened all adult patients with MMD hospitalized from January 2016 to January 2020 and collected their baseline clinical and imaging information. Univariate and multivariate logistic regression analyses were then performed to determine which imaging markers were independently associated with MMD characteristics, including cerebrovascular morphology, cerebral hemodynamics, cerebrovascular events, and postoperative collateral formation (PCF). A total of 312 cerebral hemispheres images were collected from the 156 patients with MMD. Using multivariate logistic regression analysis, the following results were generated: (1) The presence of lacunes (OR, 2.094; 95% CI, 1.109-3.955; p = 0.023) and severe white matter hyperintensities (WMH) (OR, 3.204; 95% CI, 1.742-5.892; p < 0.001) were associated with a Suzuki stage ≥ IV; (2) the presence of lacunes (OR, 6.939; 95% CI, 3.384-14.230; p < 0.001), higher numbers of enlarged perivascular spaces in centrum semiovale (CSO-EPVS) (OR, 1.046; 95% CI, 1.024-1.067; p < 0.001), and severe WMH (OR, 2.764; 95% CI, 1.463-5.223; p = 0.002) were associated with the reduced regional cerebral blood flow; (3) the presence of lacunes (OR, 12.570; 95% CI, 2.893-54.624; p = 0.001), higher numbers of CSO-EPVS (OR, 1.103; 95% CI, 1.058-1.150; p < 0.001), and severe WMH (OR, 5.982; 95% CI, 1.727-20.716; p = 0.005) were associated with ischemic cerebrovascular events; (4) the higher number of CSO-EPVS (OR, 1.077; 95% CI, 1.026-1.131; p = 0.003) was associated with good PCF. The lacunes, WMH, and CSO-EPVS were independently associated with these MMD characteristics. In conclusion, this study provided a novel and potential framework for the practical assessment of MMD by magnetic resonance imaging.