Dose-response relationship between lower serum magnesium level and higher prevalence of knee chondrocalcinosis

BACKGROUND The aim was to assess serum magnesium levels in relation to prevalence of knee chondrocalcinosis in two population-based Chinese studies. METHODS Data included in this analysis consisted of two population-based cross-sectional studies, i.e., the Xiangya Hospital Health Management Center Study and the Xiangya Osteoarthritis (XO) Study I. A bilateral knee anteroposterior radiograph was obtained from each subject. Radiographic knee chondrocalcinosis was present if there was definite linear cartilage calcification. Serum magnesium concentration was measured using the chemiluminescence method. We examined the relation of serum magnesium levels to prevalence of knee chondrocalcinosis using generalized estimating equations. RESULTS The prevalence of knee chondrocalcinosis was 1.4% in the Xiangya Hospital Health Management Center Study (n = 12,631). Compared with the lowest tertile, the age, sex and body mass index (BMI)-adjusted odds ratios (ORs) of chondrocalcinosis were 0.59 (95% CI 0.40-0.87) and 0.49 (95% CI 0.33-0.72) in the second and the third tertiles of serum magnesium, respectively (P for trend <0.001). The prevalence of knee chondrocalcinosis in the XO Study I (n = 1316) was 4.1%. The age, sex and BMI-adjusted ORs of chondrocalcinosis were 0.67 (95% CI 0.34-1.30) in the second and 0.45 (95% CI 0.21-0.94) in the third tertile of serum magnesium when compared with the lowest tertile (P for trend = 0.030). Similar results were observed in men and women in both studies. Adjusting for additional potential confounders did not change the results materially. CONCLUSIONS Subjects with lower levels of serum magnesium, even within the normal range, had higher prevalence of knee chondrocalcinosis in a dose-response relationship manner, suggesting that magnesium may have a preventive or therapeutic potential for knee chondrocalcinosis.