Cheng Chen1, Pengcheng Xun2, Frederick Unverzagt3, Leslie A McClure4, Marguerite Ryan Irvin5, Suzanne Judd6, Mary Cushman7, Ka He8,9. 1. Department of Obstetrics and Gynecology and Department of Epidemiology, Columbia University Irving Medical Center, New York, NY, USA. 2. Department of Epidemiology and Biostatistics, School of Public Health-Bloomington, Indiana University, Bloomington, IN, USA. 3. Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN, USA. 4. Department of Epidemiology and Biostatistics, Dornsife School of Public Health, Drexel University, Philadelphia, PA, USA. 5. Department of Epidemiology, School of Public Health, University of Alabama at Birmingham, Birmingham, AL, USA. 6. Department of Biostatistics, School of Public Health, University of Alabama at Birmingham, Birmingham, AL, USA. 7. Department of Medicine, Larner College of Medicine at the University of Vermont, Burlington, VT, USA. 8. Department of Obstetrics and Gynecology and Department of Epidemiology, Columbia University Irving Medical Center, New York, NY, USA. kk3399@cumc.columbia.edu. 9. Department of Obstetrics and Gynecology and Department of Epidemiology, Columbia University Irving Medical Center, 622 W 168th Street, Suite 16-62, New York, NY, 10032, USA. kk3399@cumc.columbia.edu.
Abstract
PURPOSE: To examine the prospective association between serum Mg level and the incidence of cognitive impairment. METHODS: A random sub-cohort (n = 2063) from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) cohort was included in this study. Baseline serum Mg concentration was measured using inductively coupled plasma mass spectrometry. According to the current reference interval of serum magnesium (0.75-0.95 mmol/L), we classified participants below the interval as Level 1 and used it as the referent. The rest of the study population were equally divided into three groups, named Level 2 to 4. Incident cognitive impairment was identified using the Six-Item Screener. Multivariable-adjusted odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were estimated using logistic regression models. RESULTS: After adjustment for potential confounders, an inverse threshold association between serum Mg level and incident cognitive impairment was observed. Compared to those with hypomagnesemia (Level 1: < 0.75 mmol/L), the relative odds of incident cognitive impairment was reduced by 41% in the second level [OR (95% CI) = 0.59 (0.37, 0.94)]; higher serum Mg level did not provide further benefits [Level 3 and 4 versus Level 1: OR (95% CI) = 0.54 (0.34, 0.88) and 0.59 (0.36, 0.96), P for linear trend = 0.08]. CONCLUSIONS: Findings from this prospective study suggest that sufficient Mg status within the normal range may be beneficial to cognitive health in the US general population.
PURPOSE: To examine the prospective association between serum Mg level and the incidence of cognitive impairment. METHODS: A random sub-cohort (n = 2063) from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) cohort was included in this study. Baseline serum Mg concentration was measured using inductively coupled plasma mass spectrometry. According to the current reference interval of serum magnesium (0.75-0.95 mmol/L), we classified participants below the interval as Level 1 and used it as the referent. The rest of the study population were equally divided into three groups, named Level 2 to 4. Incident cognitive impairment was identified using the Six-Item Screener. Multivariable-adjusted odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were estimated using logistic regression models. RESULTS: After adjustment for potential confounders, an inverse threshold association between serum Mg level and incident cognitive impairment was observed. Compared to those with hypomagnesemia (Level 1: < 0.75 mmol/L), the relative odds of incident cognitive impairment was reduced by 41% in the second level [OR (95% CI) = 0.59 (0.37, 0.94)]; higher serum Mg level did not provide further benefits [Level 3 and 4 versus Level 1: OR (95% CI) = 0.54 (0.34, 0.88) and 0.59 (0.36, 0.96), P for linear trend = 0.08]. CONCLUSIONS: Findings from this prospective study suggest that sufficient Mg status within the normal range may be beneficial to cognitive health in the US general population.