Kidney Week

Abstract: TH-PO214

Hypomagnesemia May Not Cause Increased Mortality as Agent Provocateur in Maintenance Hemodialysis Patients

Session Information

• Bone and Mineral Metabolism: Clinical - I
  October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Bone and Mineral Metabolism

• 402 Bone and Mineral Metabolism: Clinical

Authors

• Mizuiri, Sonoo, Division of Nephrology, Ichiyokai Harada Hospital, Hiroshima, Japan

Sonoo Mizuiri,

• Nishizawa, Yoshiko, Division of Nephrology, Ichiyokai Harada Hospital, Hiroshima, Japan

Yoshiko Nishizawa,

• Yamashita, Kazuomi, Division of Nephrology, Ichiyokai Harada Hospital, Hiroshima, Japan

Kazuomi Yamashita,

• Tanji, Chie, Ichiyokai Ichiyokai Clinic, Hiroshima, Japan

Chie Tanji,

• Usui, Kohji, Ichiyokai Ichiyokai Clinic, Hiroshima, Japan

Kohji Usui,

• Naito, Takayuki, Ichiyokai Yokokawa Clinic, Hiroshima, Japan

Takayuki Naito,

• Doi, Shigehiro, Hiroshima University Hospital, Hiroshima, Japan

Shigehiro Doi,

• Masaki, Takao, Hiroshima University Hospital, Hiroshima, Japan

Takao Masaki,

• Shigemoto, Kenichiro, Division of Nephrology, Ichiyokai Harada Hospital, Hiroshima, Japan

Kenichiro Shigemoto,

Background

It is unclear whether hypomagnesemia causes increased mortality as agent provocateur or bystander in maintenance hemodialysis (MHD) patients. This study aimed to investigate the associations between hypomagnesemia and mortality, and to study the factors related to hypomagnesemia in MHD patients.

Methods

The subjects were 353 Japanese MHD patients. Laboratory parameters including the serum magnesium (Mg), coronary artery calcium score (CACS) and medication use were studied at baseline. The subjects were stratified into baseline Mg level quartiles (Q). Kaplan-Meier survival, Cox proportional hazards models for the factors associated with mortality and logistic regression analyses for Q1 were examined.

Results

Among all patients), the median age was 68 (60-78), 39.7% were diabetics, and the median dialysis vintage was 75 (32-151) months. The Mg values of Q1 (n=86), Q2 (n=97), Q3 (n=99), and Q4 (n=71) were <2.2, 2.2-2.4, 2.5-2.7, and >2.7 mg/dl, respectively. The patients in Q1 exhibited significantly lower serum albumin, phosphate, uric acid, fibroblast growth factor 23, Kt/V_urea and normalized protein nitrogen appearance (nPNA) levels (P<0.01), but significantly higher serum high sensitivity C-reactive protein (hsCRP) levels (P<0.05) than the patients in Q2-4. During the 3 years, 91 patients died. The 3-year cumulative survival rate was 54.4%, 70.3%, 84.3%, and 85.2% in Q1, Q2, Q3, and Q4, respectively (P<0.05), and 3-year cardiovascular (CV) mortality was also significantly higher in Q1/Q2 than in Q3/Q4 (P<0.05). The Q1 was significantly associated with 3-year all-cause mortality [Hazard ratio (HR) 2.09, 95% confidence interval (CI) 0.96-4.49, P<0.01] and 3-year CV mortality [HR (95%CI): 3.03 (1.02-8.84), P<0.05], independent of Log (CACS+1), age, dialysis vintage, serum hsCRP and uric acid but the significance was lost in the model which included serum albumin, nPNA and medication use. The most significant determinant of Q1 was serum albumin levels (g/dl) [Odds ratio (95% CI): 0.10 (0.02-0.37), p<0.01], and Q1 was not found to be related to CACS or medication use including sevelamer or proton pump inhibitor.

Conclusion

Hypomagnesemia per se may not cause increased 3-year all-cause and CV mortality as agent provocateur but associated with malnutrition in MHD patients.

Funding

• Private Foundation Support