The value of mobile magnetic resonance imaging in early warning for stroke: A prospective case-control study

AIMS: To evaluate the predictive value of mobile magnetic resonance imaging (MRI) in screening stroke. METHODS: This was a prospective case-control study performed on healthy residents over 40 years old in remote rural areas of northern China between May 2019 and May 2020. Multivariate logistic regression and receiver operator characteristic curve (ROC) analysis were used to evaluate the screening model. RESULTS: A total of 1,224 patients (500 [40.8%] men) enrolled, including 56 patients who suffered from stroke (aged 64.05 ± 7.27). The individuals who developed stroke were significantly older (P < 0.001), had a significantly higher occurrence of heart disease (P = 0.015), diabetes (P = 0.005), dyslipidemia (P = 0.009), and significantly increased waist circumference (P = 0.02), systolic blood pressure (SBP) (P = 0.003), glycosylated hemoglobin (HbA1c) level (P = 0.007), triglyceride (TG) level (P = 0.025), low density lipoprotein cholesterol (LDL-c) level (P = 0.04), and homocysteine (HCY) level (P < 0.001). Multivariate logistic regression analysis showed that age (OR = 1.055, 95% CI: 1.017–1.094, P = 0.004), HCY (OR = 1.029, 95% CI: 1.012–1.047, P = 0.001) and mobile MRI (OR = 4.539, 95% CI: 1.726–11.939, P = 0.002) were independently associated with stroke. The area under the curve (AUC) of the combined model including national screening criteria, mobile MRI results, and stroke risk factors was 0.786 (95% CI: 0.721–0.851), with a sensitivity of 69.6% and specificity of 80.4%. CONCLUSION: Mobile MRI can be used as a simple and easy means to screen stroke.