Association and predictive value analysis of β2-microglobulin and the severity of white matter hyperintensities

Abstract Objective: To investigate the association between serum β2-microglobulin (β2M) levels and the severity of White Matter Hyperintensities (WMHs) in patients with Cerebral Small Vessel Disease (CSVD), in addition to evaluate its predictive value for WMHs severity. Methods: We consecutively enrolled in patients with CSVD demonstrating WMHs on MRI from the Neurology Department at the Second People's Hospital of Hefei City between December 2021 and April 2024. Patient characteristics including demographic,baseline clinical, laboratory data, serum β2M levels and brain MRI features were collected. The severity of periventricular white matter hyperintensities (PWMHs) and deep white matter hyperintensities (DWMHs) was assessed using the Fazekas scale. Based on the sum of the scores from these two regions, patients were classified into a none or mild overall WMHs group (Fazekas score 0–2) and a moderate to severe overall WMHs group (Fazekas score 3–6). Patients were classified into a predominant periventricular white matter hyperintensities (pred-PWMHs) subgroup and a predominant deep white matter hyperintensities (pred-DWMHs) subgroup based on a score difference of ≥ 1 point between the two regions. Each subgroup was further divided into mild (Fazekas score 1) and moderate to severe groups (Fazekas score 2–3). Independent risk factors associated with moderate to severe overall WMHs, PWMHs and DWMHs in patients with CSVD were analyzed using univariate and multivariate logistic regression. The predictive value of β2M for moderate to severe overall WMHs was evaluated using receiver operating characteristic (ROC) curves. Results: A total of 346 patients were enrolled in the study, including 183 patients with none or mild overall WMHs and 163 patients with moderate to severe overall WMHs. Univariate analysis revealed that age (P < 0.001), hypertension (P < 0.001), fibrinogen (P = 0.019), Hcyt(P = 0.005), CysC (P = 0.001), Total cholesterol (P = 0.025), LDL-C(P = 0.012), eGFR (P < 0.001) and β2M (P < 0.001) were associated with the severity of overall WMHs. Multivariate logistic regression analysis identified age (OR:1.050, 95%CI:1.025–1.075, P < 0.001༉, hypertension (OR:2.007, 95%CI:1.202–3.349, P = 0.008) and β2M (OR:1.635, 95%CI:1.154–2.317, P = 0.006) as independent risk factors for moderate to severe overall WMHs. ROC curve analysis demonstrated that a β2M cut off value of 2.295 was significantly predictive of moderate to severe overall WMHs (AUC = 0.673, P < 0.001).In subgroup analysis, β2M was also identified as an independent risk factor for moderate to severe pred-PWMHs (OR:3.134, 95%CI: 1.012–9.698, P = 0.048),while no association was observed with the severity of pred-DWMHs. Conclusion: Serum β2M levels are significantly associated with the severity of overall WMHs and pred-PWMHs,but not pred-DWMHs. Furthermore, β2M levels exhibit predictive value for moderate to severe overall WMHs.