Magnesium and Dialysis: The Neglected Cation

In Practice Magnesium and Dialysis: The Neglected Cation Mohamad Alhosaini, MD,1,2 and David J. Leehey, MD1,2 Disorders of magnesium homeostasis are very common in dialysis patients but have received scant attention. In this review, we address measurement of plasma magnesium, magnesium balance and the factors that affect magnesium flux during dialysis, the prevalence of hypo- and hypermagnesemia in dialysis patients, and the potential clinical significance of hypo- and hypermagnesemia in dialysis patients. Many factors can affect plasma magnesium concentration, including diet, nutritional status (including plasma albumin level), medications (such as proton pump inhibitors), and dialysis prescription. Further interventional studies to determine the effect of normalization of plasma magnesium concentration on clinical outcomes are needed. At the present time, we recommend that predialysis plasma magnesium be measured on a regular basis, with the dialysate magnesium concentration adjusted to maintain plasma magnesium concentration within the normal range. Am J Kidney Dis. -(-):---. ª 2015 by the National Kidney Foundation, Inc. INDEX WORDS: Magnesium; dialysis; hypomagnesemia; hypermagnesemia; proton pump inhibitors; hemodialysis; peritoneal dialysis; Gibbs-Donnan effect; review. Note from Editors: This article is part 3 of a 4-part series of invited In Practice reviews highlighting issues related to the composition of dialysate. CASE PRESENTATION A 50-year-old man with end-stage kidney disease due to diabetes mellitus on thrice-weekly hemodialysis (HD) therapy for the past 5 years was seen urgently in the dialysis unit because of an abnormal heart rhythm. His medical history was remarkable for diabetes, hypertension, coronary artery disease, and chronic dyspepsia of unclear cause. Medications included insulin, lisinopril, calcitriol, epoetin alfa, calcium acetate, and pantoprazole, all of which he had been taking since the onset of maintenance HD therapy. His dialysis prescription used the following dialysate composition: sodium, 140 mEq/L; potassium, 2 mEq/L; bicarbonate, 35 mEq/L; acetate, 4 mEq/L; calcium, 2.5 mEq/L; and magnesium, 0.75 mEq/L. On examination, his heart rate was regular but rapid (160 beats/min). An electrocardiogram showed the ventricular arrhythmia torsades de pointes. He underwent cardioversion, after which he developed a junctional escape rhythm. Review of his monthly laboratory values drawn before the previous dialysis session showed the following: sodium, 140 mEq/L; potassium, 5 mEq/L; chloride, 102 mEq/L; bicarbonate, 22 mEq/L; total calcium, 9 mg/dL; and inorganic phosphorus, 5 mg/dL. These values were similar to his previous monthly values. Stat plasma magnesium concentration was 1.0 (reference range, 1.7-2.4) mg/dL. Dialysis was discontinued and magnesium sulfate, 2 g, was administered intravenously over 5 minutes, followed by an additional 2 g over 1 hour, with prompt restoration of sinus rhythm. He was admitted to the hospital telemetry unit for observation. Repeat plasma magnesium concentration 4 hours after the end of the magnesium infusion was 1.5 mg/dL and an additional 2 g of magnesium sulfate was administered intravenously over 4 hours. Pantoprazole therapy was discontinued. After discharge, dialysate magnesium concentration was increased to 1.0 mEq/L, after which predialysis plasma magnesium concentration was maintained in the normal range. He had no recurrence of arrhythmia. INTRODUCTION Magnesium is the fourth most abundant cation in the human body and the second most abundant in the intracellular space. However, magnesium receives only scant attention from most clinicians caring for dialysis patients. It has been long referred to as “the neglected cation.”1 This may be due to the controversies about the clinical significance of disorders of magnesium homeostasis and the true risks of hypoand hypermagnesemia. In the early days of dialysis, there were many publications dealing with magnesium level abnormalities; after a long fallow period, there has been some recent resurgence of interest in the area, especially regarding the possible relationship between hypomagnesemia and cardiovascular disease. In this review, we address measurement of plasma magnesium, magnesium balance and factors that affect magnesium flux during dialysis, the prevalence of hypo- and hypermagnesemia in dialysis patients, and the potential clinical significance of hypo- and hypermagnesemia in dialysis patients. MEASUREMENT OF MAGNESIUM IN BODY FLUIDS Normal plasma (or serum) total magnesium concentration is 0.7 to 1.0 mmol/L. Because magnesium is a divalent ion with an atomic mass of 24.305, this translates to 1.4 to 2.0 mEq/L or 1.7 to 2.4 mg/dL (Fig 1). The normal concentration in cells has been reported to be 5 to 20 mmol/L. Extracellular magnesium accounts for only w1% of total-body magnesium From 1Loyola University Medical Center, Maywood; and Veterans Affairs Hospital, Hines, IL. Received September 9, 2014. Accepted in revised form January 13, 2015. Address correspondence to David J. Leehey, MD, Hines VA Hospital (111-L), Hines, IL 60141. E-mail: dleehey@lumc.edu  2015 by the National Kidney Foundation, Inc. 0272-6386 http://dx.doi.org/10.1053/j.ajkd.2015.01.029 2 Am J Kidney Dis. 2015;-(-):--FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce 1 web 4C=FPO Alhosaini and Leehey Figure 1. Normal plasma (or serum) magnesium concentration. Conversion factors are given in the boxes. To convert from mg/dL to mEq/L, divide by 1.2. To convert from mEq/L to mmol/ L, divide by 2. To convert from mg/dL to mmol/L, divide by 2.4. For the opposite conversions (eg, mEq/L to mg/dL), multiply by the corresponding factors. Conversion factors are derived as follows. Since for magnesium (Mg), 1 mmol w 24 mg, to convert from mg/dL to mmol/L, divide by one-tenth of the atomic weight, that is, 0.1 3 24, or 2.4. To convert mmol/L to mEq/L, multiply by 2, because Mg is a divalent ion. To directly convert mg/dL to mEq/L, divide by 2.4/2 5 1.2. As an example, if the plasma Mg concentration ([Mg]) is 1.7 mg/dL, this is equivalent to (1.7/ 2.4) 3 2 5 1.7/1.2 5 1.4 mEq/L. content. About 60% to 70% of plasma magnesium is in the free or ionized form, with w25% bound to proteins, predominantly albumin, and the remaining 5% to 10% complexed with nonprotein anions such as bicarbonate, phosphate, and citrate (although the percentage of complexed magnesium increases in endstage renal disease and has been reported to be as high as 16% in patients on continuous ambulatory peritoneal dialysis therapy2). Ionized magnesium and complexed magnesium together form the ultrafilterable fraction of magnesium.3 This ultrafilterable fraction represents the portion of total plasma magnesium that can be removed by the kidneys or dialysis. Measurement of plasma (or serum) total magnesium is generally performed by dye-based spectrophotometric (colorimetric) methods and occasionally by other techniques such as enzymology, fluorometry, flame-emission spectrometry, or atomic absorption spectrometry.3 A commonly used dye-based spectrophotometric assay uses calmagite and methylthymol blue.4 These substances form a colored complex with magnesium in alkaline solution, which is measured at w600 nm. Specific calcium chelating agents such as ethylene glycol tetraacetic acid (EGTA) are added to reduce interference by calcium. Free (ionized) magnesium concentration can be determined by commercially available instruments using ion-selective electrodes with neutral carrier ionophores. However, there is a problem of cross-reactivity with calcium, so ionized calcium must be simultaneously determined and subtracted in order to calculate free magnesium concentration. Because most studies have shown good correlation between total and ionized plasma magnesium concentrations, the latter is not widely used in clinical practice. However, in the presence of severe hypoalbuminemia, total plasma magnesium concentration may be slightly low despite a normal plasma ionized magnesium concentration due to a decrease in 2 the protein-bound fraction. Intracellular magnesium concentration is generally measured by fluorescent probes or magnetic resonance methods and is a research tool. The definition of normal plasma (or serum) total magnesium concentration (generally referred to as simply plasma [or serum] magnesium concentration, because ionized magnesium is not commonly measured in clinical laboratories) is somewhat controversial. US laboratories generally report magnesium concentrations as milligrams per deciliter; the lower limit used by most clinical laboratories is between 1.6 and 1.8 mg/dL, while the upper limit is usually between 2.2 and 2.4 mg/dL. The largest study in which plasma magnesium was measured was done in the 1970s involving 15,820 healthy adults.5 Mean serum magnesium concentration was 2.0 mg/dL, with 95% of values between 1.8 and 2.2 mg/dL. However, this range may reflect magnesium intake in the modern time and may not necessarily represent the true normal range.6 It is believed that the introduction of processed food in the 20th century led to a decrease in magnesium intake compared to that of previous centuries. Suggestions have thus been made to increase the lower limit of normal plasma magnesium concentration to 2.0 mg/dL. In support of this view is the fact that many observational studies have shown that the lowest risk for diabetes mellitus and cardiovascular mortality was in the range of 1.9 to 2.7 mg/dL.7-11 MAGNESIUM BALANCE In a healthy individual, total-body magnesium content is kept constant by interactions among intestine, bones, and the kidneys (Fig 2). When magnesium intake is low, the intestine can increase the amount of dietary magnesium absorbed from 40% to 80%,12 and urinary fractional excretion of magnesium can be decreased to ,0.5%.13 The increase in intestinal absorption occurs by passive paracellular transport, as well as by active membrane-bound channels (discussed next). If magnesium intake continues to be low, bones, which have w55% of total-body magnesium, will slowly release magnesium to the plasma. In the case of high magnesium intake, healthy kidneys are capable of increasing urinary magnesium excretion to keep plasma magnesium concentration within the normal range. Unlike other ions, there is limited hormonal regulation of magnesium balance. Active vitamin D can increase intestinal magnesium absorption.14 Epidermal growth factor and estrogens increase distal tubule magnesium reabsorption,15 though the clinical significance of this is unclear. In malabsorptive states, magnesium absorption is known to be affected by free fatty acids in the intestinal lumen, which may combine with magnesium to form nonabsorbable soaps (saponification). Sevelamer Am J Kidney Dis. 2015;-(-):--- FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce web 4C=FPO Magnesium and Dialysis Figure 2. Normal magnesium balance. Two pathways are responsible for magnesium absorption in the intestine: a passive paracellular pathway and an active transcellular pathway. The passive pathway is mainly in the distal jejunum and ileum and the active pathway is thought to be in the ileum and colon. The passive pathway is responsible for 80% to 90% of intestinal magnesium absorption. It is a nonsaturable process and depends on concentration and electrical gradients between the lumen and enterocyte cytoplasm. Claudin 16 and 19 probably play a major role in this process. Transient receptor potential melastatin (TRPM) 6 and 7 are the 2 transporters that are involved in the active pathway, which is responsible for the other 10% to 20% of intestinal magnesium absorption. The active pathway is upregulated when magnesium intake is low. In the kidney, 90% to 95% of filtered magnesium is reabsorbed. Unlike other ions, only small amounts of magnesium (10%-25%) are reabsorbed in the proximal tubule, with 70% reabsorbed through the paracellular pathway in the loop of Henle. Claudin 16 and 19 may also play a role in this process. About 10% of magnesium is then reabsorbed in the distal tubule through an active process mediated by the TRPM 6 channel. (Figure reproduced from de Baaij et al.70 Regulation of magnesium balance: lessons learned from human genetic disease. Clin Kidney J. 2012;5(suppl 1):i15-i24.) hydrochloride use is associated with an increase in plasma magnesium concentration; this effect may be mediated by binding of sevelamer to bile salts, thus increasing the amount of free magnesium available for absorption.16 Recent research suggests that certain dietary fibers enhance mineral absorption, including magnesium.17 Studies in rats have demonstrated that net absorption of magnesium in vivo is depressed by high calcium intake, although the importance of this interaction has been questioned in studies of humans. Moreover, the mechanism by which calcium and magnesium interact has not been well defined.18 It is unlikely that calcium-containing phosphate binders have an important effect on magnesium absorption, but this has not been well studied. Transient receptor potential channel melastatin member 6 (TRPM6) and TRPM7 are newly discovered channels that are involved in active magnesium transport. TRPM6 is part of the transient receptor potential family of cation channels. It is permeable to both magnesium and calcium, but has 5 times more affinity to magnesium than calcium. It is expressed in the distal tubule and the small intestine brush border membrane, where it serves to increase intestinal magnesium uptake in the face of low magnesium intake. Mutations of TRPM6 cause rare genetic disorders characterized by hypomagnesemia and hypocalcemia.19,20 Epidermal growth factor and estrogen activate the transcription of TRPM6. Changes in intestinal pH caused by proton pump inhibitors (PPIs), but not antihistamine receptor type 2 blockers, decrease the activity of intestinal TRPM6 and thus predispose to hypomagnesemia in patients with low dietary magnesium intake and/or increased losses into urine.21 In dialysis patients, losing the regulatory role of the kidneys can have significant effects on magnesium balance. It might appear that hypermagnesemia is the only possible outcome in such patients. In the early days of dialysis, there was concern about hypermagnesemia and little if any concern about hypomagnesemia. However, dialysis patients in the modern era are usually normomagnesemic and sometimes even hypomagnesemic.22,23 This is because of complex effects of magnesium intake, other dietary intake, drugs, and dialysate magnesium concentration on magnesium balance and thus total-body magnesium content. For example, PPIs are commonly used by dialysis patients. These drugs impair the adaptive increase in active intestinal magnesium absorption in the face of magnesium depletion, predisposing to hypomagnesemia.22 Use of low-magnesium dialysate is a risk factor for hypomagnesemia in both HD and continuous ambulatory peritoneal dialysis patients.22,23 Patients with chronic kidney disease normally have severely depressed intestinal magnesium absorption compared with healthy individuals,24 probably due to a deficiency of active vitamin D. Thus, although there is minimal renal magnesium excretion, total-body magnesium content can be low, normal, or high in dialysis patients. Some of these factors are discussed in more detail later in this article. MAGNESIUM FLUX DURING DIALYSIS The 2 major factors that affect magnesium diffusion during dialysis are the concentration gradient across the dialysis membrane and the Gibbs-Donnan effect. Ultrafilterable magnesium (ionized plus complexed magnesium) is the only portion of total plasma magnesium that can be removed by dialysis or ultrafiltration. Albumin and several nonprotein anions bind to magnesium, and because plasma concentrations of these are variable, ultrafilterable magnesium concentration also is somewhat variable, but usually is w70% of total plasma magnesium concentration. With HD, the concentration gradient, that is, the difference between the ultrafilterable plasma magnesium Am J Kidney Dis. 2015;-(-):--FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce 3 Alhosaini and Leehey concentration and dialysate magnesium concentration, is the primary determinant of magnesium flux, with magnesium being removed if ultrafilterable plasma magnesium concentration exceeds dialysate magnesium concentration. A small amount of magnesium is also removed by ultrafiltration. With peritoneal dialysis, increasing ultrafiltration by the use of more hypertonic dialysis solutions substantially increases magnesium loss into the dialysate.25 The other factor to consider when predicting magnesium diffusion during dialysis is the Gibbs-Donnan effect, in which ion transport across a membrane is influenced by the unequal distribution of proteins between plasma and dialysate.2 Anionic proteins such as albumin will impede movement of cations such as magnesium across the membrane. For a monovalent ion, the Donnan factor is 0.96, and for a divalent ion, it is 0.962, or 0.92. Taking into account the Donnan factor, one might predict that dialysate magnesium concentration would need to be w8% lower than the ultrafilterable plasma magnesium concentration in order for there to be equilibrium between dialysate and plasma. However, it also must be remembered that due to the volume occupied by proteins and lipids in plasma, plasma water concentrations of ions are w5% higher than total plasma concentrations. When these factors are taken together, the equilibrium dialysate concentration, that is, the dialysate magnesium concentration at which there is no net transport of magnesium across the membrane, will be w3% lower in dialysate than plasma. Concentrations of magnesium used in dialysate have changed over the years. In the 1970s and 1980s, it was more common to use concentrations of w1.5 mEq/L, but more recently, lower concentrations are generally used (the current standard in the United States is 0.751.0 mEq/L). The reason behind this change is not clear, but most likely is multifactorial. The introduction of magnesium carbonate as a phosphate binder26 and the relief of itching27 and bone disease28 in a few studies when lower dialysate magnesium was used are contributing factors. Although magnesium-containing phosphate binders are rarely prescribed at the present time, lower dialysate magnesium concentration remains in widespread use, with little or no attention given to plasma magnesium concentration in most dialysis units in the United States. In HD, with a 1.5-mEq/L dialysate magnesium concentration, most patients will have normal or slightly high serum magnesium concentrations.29,30 Use of a 1.0-mEq/L dialysate magnesium concentration usually results in normal predialysis plasma magnesium concentrations, but at least 5% of patients will have predialysis hypomagnesemia.31,32 Hypomagnesemia is even more common with a 0.5-mEq/L dialysate magnesium concentration (5%-33%), and high plasma magnesium concentration 4 is much less common.29,31 In a recent study, with use of a 0.75- to 1.0-mEq/L dialysate magnesium concentration, hypomagnesemia was found to be a common occurrence (seen in 38% of patients), but many of these patients were taking a PPI, which can prevent magnesium absorption by the intestine.22 In peritoneal dialysis, the same pattern is observed; most patients have normal or high plasma magnesium concentrations with a 1.5-mEq/L dialysate,33-37 but hypomagnesemia is common when a 0.5-mEq/L dialysate magnesium concentration is used.23,37,38 Hypoalbuminemia will decrease the Donnan effect and thus tend to increase diffusion of magnesium into dialysate. In addition, the fraction of total magnesium that is protein bound is decreased and the concentration of ultrafilterable magnesium is increased by hypoalbuminemia, which will also tend to increase magnesium removal. For example, if a dialysate magnesium concentration of 1.0 mEq/L is used and assuming that two-thirds of plasma total magnesium concentration is ultrafilterable, little magnesium should be removed in a patient with a plasma total magnesium concentration of 1.5 mEq/ L. However, in the presence of hypoalbuminemia, at the same dialysate magnesium concentration and plasma total magnesium concentration, greater magnesium elimination will occur by 2 mechanisms: higher free plasma magnesium and lower GibbsDonnan effect. If a dialysate magnesium concentration , 1.0 mEq/L is used, net removal of magnesium will occur, especially in hypoalbuminemic patients (Fig 3). In addition, other facets of the dialysis prescription may affect magnesium flux across the dialyzer and/or plasma magnesium concentration. An increase in dialysate bicarbonate concentration and thus blood pH will increase the number of anionic sites on albumin and increase magnesium binding to albumin in plasma, lowering the plasma ionized magnesium fraction. The magnitude of the decrease in ionized magnesium concentration with an increase in pH has been reported to be 0.12 mmol/L (0.29 mg/dL) per pH unit, which is significantly less than that of ionized calcium (0.36 mmol/L per pH unit) and thus important only at extremes of the physiologic pH range.39 However, use of citrate-containing dialysate would be expected to increase the concentration of magnesium complexed to citrate in plasma and decrease protein-bound magnesium; because citrate-magnesium complexes are dialyzable, this would be expected to increase dialytic magnesium removal. It is unlikely that bicarbonate would have such an effect because the existence of magnesium-bicarbonate complexes in plasma has been questioned.40 Glucose administered in dialysate will stimulate insulin. Insulin increases cellular magnesium uptake into insulin-sensitive tissues,41,42 and an increase Am J Kidney Dis. 2015;-(-):--- FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce web 4C=FPO Magnesium and Dialysis Figure 3. Factors affecting dialysate magnesium (Mg) losses. (Right panel) Little Mg is removed with dialysis because the GibbsDonnan effect substantially counterbalances the concentration gradient between plasma dialyzable Mg and dialysate Mg. (Left panel) However, hypoalbuminemia will lessen the Gibbs-Donnan effect and there will be relatively more Mg removed by dialysis. In addition, hypoalbuminemia increases the ratio of dialyzable to total plasma Mg, resulting in more dialytic Mg removal at any total plasma Mg concentration ([Mg]). in insulin resulting from hyperglycemia could enhance the effects of a low plasma magnesium concentration. Other factors that could affect magnesium removal other than dialysate magnesium composition include blood flow rate and dialysate flow rate, although to our knowledge there have been no studies of these factors. With current US dialysis practice, in which a dialysate magnesium concentration of 0.75 to 1.0 mEq/L is typical, magnesium will be lost into the dialysate and over time, magnesium depletion may develop, especially in patients with low magnesium intake and/or who are taking a PPI. A summary of the effect of different dialysate magnesium concentrations on serum total magnesium concentrations in dialysis patients is given in Table 1. ACUTE EFFECTS OF LOW-MAGNESIUM DIALYSATE Use of a low-magnesium dialysate may have acute hemodynamic effects during HD. In a study from Greece, 14 dialysis patients with a fixed dialysate calcium concentration were studied on 3 different dialysate magnesium concentrations (0.5 vs 1.0 vs 1.5 mEq/L). Episodes of hypotension during dialysis were least common with the highest magnesium dialysate and most common with the lowest magnesium dialysate. The dialysate magnesium concentration of 0.5 mEq/L was associated with lower cardiac output but the same peripheral vascular resistance and heart rate compared with higher magnesium dialysates, suggesting that the adverse effects of lowmagnesium dialysate on blood pressure is probably related to impaired myocardial contractility.43 Another study from Egypt also showed a positive correlation between hypotension and a decrease in serum magnesium concentration during dialysis.44 Finally, in a study from Iran, the occurrence of intradialytic hypotension was also found to be significantly related to a decrease in serum magnesium concentration during HD.45 CLINICAL SIGNIFICANCE OF CHRONIC HYPOMAGNESEMIA AND CHRONIC HYPERMAGNESEMIA IN DIALYSIS PATIENTS Cardiovascular Disease During the past decade, there has been increased interest in the effects of magnesium on cardiovascular diseases. In particular, the effects of magnesium disorders on arrhythmias and vascular calcification are of great clinical relevance. A direct relationship between magnesium disorders and arrhythmias in dialysis patients remains unclear. Magnesium has complex effects on myocardial ion fluxes. Magnesium deficiency impairs sodiumpotassium adenosine triphosphatase, leading to a decrease in intracellular potassium and a relatively depolarized resting membrane potential, predisposing to arrhythmia. In nondialysis patients, there are anecdotal reports of monomorphic ventricular tachycardia, torsades de pointes, and ventricular fibrillation associated with hypomagnesemia that responded to magnesium repletion.46,47 Concomitant potassium depletion further increases the risk, and use of a low dialysate potassium concentration may further increase the risk of low dialysate magnesium concentration, though this has not been studied. However, magnesium excess may also be deleterious. A 50-ms difference in corrected QT (QTc) interval dispersion is associated with 1.5-fold increase in mortality in dialysis patients.48 The relationship between magnesium and QTc dispersion was tested in a pilot study in which 18 dialysis patients were placed on high oral magnesium intake (500-700 mg/d) and 12 patients were placed on lower magnesium intake (400-500 mg/d).49 Peripheral-blood mononuclear cell magnesium content was measured at baseline and after 6 months. The higher oral magnesium group had substantially higher peripheral-blood mononuclear cell magnesium content compared to the lower magnesium group. QTc dispersion was significantly higher in the Am J Kidney Dis. 2015;-(-):--FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce 5 Alhosaini and Leehey Table 1. Effect of Dialysate Magnesium Concentration on Serum Total Magnesium Concentration Study Dialysate Magnesium Concentration (mEq/L) N Hemodialysis 13 1.5 0.5 12 22a 1.5 Nilsson et al30 (1984) 0.4 Saha et al31 (1996) 47a 1.0 0.5 1.0 Navarro et al32 (1999) 110 Alhosaini et al22 (2014) 62 0.75-1.0 Gonella et al29 (1981) Blumenkrantz et al33 (1982) Ejaz et al23 (1995) Hutchison et al34 Peritoneal Dialysis 8 1.5 21 12 (1996) 11a Saha et al35 (1997) Eisenman et al37 13 10 3 (2003) 19 15 (2005) 5a Ye et al38 (2013) 402 Katopodis et al36 0.5 0.5 1.5 0.5 1.5 1.0 0.5 1.5 1.0 1.5 0.5 0.5 Serum Total Magnesium Concentration (mEq/L) 2.34 6 0.10 1.56 6 0.04 2.36 6 0.26 1.88 6 0.48 2.02 6 0.38 1.88 6 0.36 2.30 6 0.33 1.48 6 0.16 (PPI)b; 1.65 6 0.26 (non-PPI)b 2.54 6 0.08 1.10 6 0.02c 1.60 6 0.07c 2.48 6 0.12 1.88 6 0.08 2.44 6 0.40 2.04 6 0.30 1.96 6 0.70 2.30 6 0.22 1.88 6 0.28 1.69 6 0.26 1.42 6 0.39 1.46 6 0.22 Abbreviation: PPI, proton pump inhibitor. Total patients, including both dialysate concentration groups. b Mean values for patients taking and not taking PPIs. c Mean values for hypo- and normomagnesemic patients. a higher magnesium group compared to the lower magnesium group, suggesting a possible adverse effect of high intracellular magnesium content on the cardiac conduction system. Severe hypermagnesemia is known to cause cardiac conduction defects, including bradyarrhythmias, complete heart block, and neuromuscular effects such as loss of deep tendon reflexes and muscle weakness.50 However, these findings are rarely seen unless plasma magnesium concentration is elevated (.45 mg/dL), which is now very uncommon in HD patients. There may be a long-term protective effect of higher serum magnesium concentration due to its inhibitory role in vascular calcification. In vitro and animal studies have shown that magnesium could inhibit vascular calcifications through 2 mechanisms: decreasing transformation of calcium phosphate nanocrystals to a stable apatite51 and decreasing transformation of vascular smooth myocytes into osteoblasts.52 Inhibition of Wnt/b-catenin signaling by magnesium is one potential intracellular mechanism by which this anticalcifying effect is achieved.53 In humans, several small interventional studies give support to these experimental studies. Coronary artery calcification 6 score progression on computed tomography was slower in 7 dialysis patients who were placed on oral calcium and magnesium carbonate compared to a historical control group.54 In comparison to calcium acetate alone, the combination of magnesium citrate and calcium acetate increased serum magnesium concentration and reduced carotid intima-media thickness.55 Finally, orally administered magnesium carbonate as a phosphate binder may retard the progression of arterial calcifications in HD patients.56 Bone Disease The inhibitory effects of magnesium on calcification are not restricted to blood vessels, but are also seen in bone, where magnesium may inhibit mineralization. Hypermagnesemia has been linked to osteomalacia. In a pilot study, 10 dialysis patients were switched from a dialysate magnesium concentration of 1.0 mEq/L to 0.5 mEq/L for 1 year. Repeat bone biopsy showed reduction in the degree of osteomalacia.28 Magnesium also activates the calcium-sensing receptor and thus suppresses parathyroid hormone (PTH) synthesis and secretion. Hypermagnesemia is associated with lower PTH levels,32,57 whereas lower dialysate magnesium concentration in peritoneal dialysis58,59 and HD patients60,61 is associated with higher PTH levels. Serum magnesium concentration is inversely related to both PTH62 and fibroblast growth factor 23 (FGF-23) levels63 in HD patients. In studies in which rat parathyroid glands were incubated in different calcium and magnesium concentrations, magnesium was able to reduce PTH level only in the presence of low calcium concentrations; with normal-high calcium concentrations, the effect of magnesium on PTH inhibition was minor or absent.64 Animal studies have shown that magnesium deficiency leads to increased substance P, tumor necrosis factor a, interleukin 1b, and receptor activator of nuclear factor-kB ligand (RANKL) in bone, which increases osteoclast activity and bone resorption.65 These associations raise the concern of adynamic bone disease in hypermagnesemic patients and osteitis fibrosa in hypomagnesemic patients. Moreover, there is epidemiologic evidence of an association between magnesium deficiency and osteoporosis in the general population.66 Whether this is true for dialysis patients is unknown. Looking at the overall picture, bone disease in dialysis patients is a complex process that involves calcium, phosphorus, PTH, vitamin D, FGF-23, metabolic acidosis, and medications. Magnesium represents only 0.5% of bone ash, and there is no simple association between magnesium disorders and a specific bone disease pattern. Mortality Several large observational studies relating low blood magnesium concentration to mortality have been Am J Kidney Dis. 2015;-(-):--- FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce Magnesium and Dialysis recently reported.9,11,67 The largest study came from Japan in 2014.11 In this study, serum magnesium concentrations of 142,555 dialysis patients were available from a large nationwide database. After 1 year, a fully adjusted model showed a J-shaped association between serum magnesium concentration and cardiovascular mortality. Participants were divided into 6 sextiles of serum magnesium concentration (,2.3, 2.3-2.5, 2.5-2.7, 2.7-2.8, 2.8-3.1, and .3.1 mg/dL). Mortality was highest with serum magnesium concentrations , 2.7 and . 3.1 mg/dL. Hypomagnesemia was a significant predictor of not only cardiovascular, but also noncardiovascular mortality. Hypermagnesemia was also associated with higher mortality, but the magnitude of the effect was much less than with hypomagnesemia. In an accompanying editorial, it was noted that patients with the lowest serum magnesium values tended to be older, have worse nutritional status, and have more comorbid conditions, pointing to the possibility of magnesium being an innocent bystander rather than an agent provocateur,68 and the editorial recommended further interventional trials. RECOMMENDATIONS We recommend that predialysis plasma magnesium be measured on a regular basis, with dialysate magnesium concentration adjusted to maintain plasma magnesium concentration within the normal range. This is particularly important in malnourished patients and patients taking PPIs. Dialysate magnesium concentration should generally be 1.0 mEq/L, but a minority of patients will require a dialysate magnesium concentration of 1.25 to 1.5 mEq/L. A dialysate magnesium concentration , 1.0 mEq/L should be used sparingly, if at all. SUMMARY Although there has been a resurgence of interest in magnesium in recent years, it still unfortunately remains the neglected cation.1 Magnesium is essential for health and is involved in a great number of crucial physiologic functions. Moreover, changes in dietary habits and in particular the steadily increasing use of refined rather than whole foods is likely contributing to an epidemic of magnesium deficiency. Dialysis patients may be particularly vulnerable to the effects of magnesium deficiency, yet scant if any attention is being given to this cation in most dialysis centers, which should be rectified. Perhaps it is time to return to the basics.69 ACKNOWLEDGEMENTS Support: None. Financial Disclosure: The authors declare that they have no relevant financial interests. REFERENCES 1. Parsons FM, Stewart WK. The composition of dialysis fluid. In: Drukker W, Parsons FM, Maher JF, eds. Replacement of Renal Function by Dialysis. 2nd ed. Boston, MA: Martinus Nijhoff Publishers; 1983:155. 2. Rippe B, Venturoli D. Optimum electrolyte composition of a dialysis solution. Perit Dial Int. 2008;28(suppl 3):S131-S136. 3. Huijgen HJ, Van Ingen HE, Kok WT, Sanders GTB. Magnesium fractions in the serum of healthy individuals and CAPD patients, measured by an ion-selective electrode and ultrafiltration. Clin Biochem. 1996;29(3):261-266. 4. Endres DB, Rude RK. Disorders of bone. In: Burtis CA, Ashwood ER, Bruns DE, Sawyer BG, eds. Tietz Fundamentals of Clinical Chemistry. 6th ed. St Louis, MO: Saunders Elsevier; 2008:719-721. 5. Lowenstein FW, Stanton MF. Serum magnesium levels in the United States, 1971-1974. J Am Coll Nutr. 1986;5(4):399-414. 6. Ford ES, Mokdad AH. Dietary magnesium intake in a national sample of US adults. J Nutr. 2003;133(9):2879-2882. 7. Everett CJ, King DE. Serum magnesium and the development of diabetes. Nutrition. 2006;22(6):679. 8. Guerrero-Romero F, Rascón-Pacheco RA, RodríguezMorán M, de la Peña JE, Wacher N. Hypomagnesaemia and risk for metabolic glucose disorders: a 10-year follow-up study. Eur J Clin Invest. 2008;38(6):389-396. 9. Ishimura E, Okuno S, Yamakawa T, Inaba M, Nishizawa Y. Serum magnesium concentration is a significant predictor of mortality in maintenance hemodialysis patients. Magnes Res. 2007;20(4):237-244. 10. Kanbay M, Yilmaz MI, Apetrii M, et al. Relationship between serum magnesium levels and cardiovascular events in chronic kidney disease patients. Am J Nephrol. 2012;36(3):228-237. 11. Sakaguchi Y, Fujii N, Shoji T, Hayashi T, Rakugi H, Isaka Y. Hypomagnesemia is a significant predictor of cardiovascular and non-cardiovascular mortality in patients undergoing hemodialysis. Kidney Int. 2014;85(1):174-181. 12. Graham LA, Caesar JJ, Burgen AS. Gastrointestinal absorption and excretion of Mg 28 in man. Metabolism. 1960;9:646-659. 13. Topf JM, Murray PT. Hypomagnesemia and hypermagnesemia. Rev Endocr Metab Disord. 2003;4(2):195-206. 14. Schmulen AC, Lerman M, Pak CY, et al. Effect of 1,25(OH)2D3 on jejunal absorption of magnesium in patients with chronic renal disease. Am J Physiol. 1980;238(4):G349-G352. 15. Groenestege WM, Hoenderop JG, van den Heuvel L, Knoers N, Bindels RJ. The epithelial Mg21 channel transient receptor potential melastatin 6 is regulated by dietary Mg21 content and estrogens. J Am Soc Nephrol. 2006;17(4):1035-1043. 16. Mitsopoulos E, Griveas I, Zanos S, et al. Increase in serum magnesium level in haemodialysis patients receiving sevelamer hydrochloride. Int Urol Nephrol. 2005;37(2):321-328. 17. Legette LL, Lee W, Martin BR, Story JA, Campbell JK, Weaver CM. Prebiotics enhance magnesium absorption and inulin-based fibers exert chronic effects on calcium utilization in a postmenopausal rodent model. J Food Sci. 2012;77(4): H88-H94. 18. Hardwick LL, Jones MR, Brautbar N, Lee DB. Magnesium absorption: mechanisms and the influence of vitamin D, calcium and phosphate. J Nutr. 1991;121(1):13-23. 19. Walder RY, Landau D, Meyer P, et al. Mutation of TRPM6 causes familial hypomagnesemia with secondary hypocalcemia. Nat Genet. 2002;31(2):171-174. 20. Schlingmann KP, Weber S, Peters M, et al. Hypomagnesemia with secondary hypocalcemia is caused by mutations in Am J Kidney Dis. 2015;-(-):--FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce 7 Alhosaini and Leehey TRPM6, a new member of the TRPM gene family. Nat Genet. 2002;31(2):166-170. 21. Bai JP, Hausman E, Lionberger R, Zhang X. Modeling and simulation of the effect of proton pump inhibitors on magnesium homeostasis. 1. Oral absorption of magnesium. Mol Pharm. 2012;9(12):3495-3505. 22. Alhosaini M, Walter JS, Singh S, Dieter RS, Hsieh A, Leehey DJ. Hypomagnesemia in hemodialysis patients: role of proton pump inhibitors. Am J Nephrol. 2014;39(3):204-209. 23. Ejaz AA, McShane AP, Gandhi VC, Leehey DJ, Ing TS. Hypomagnesemia in continuous ambulatory peritoneal dialysis patients dialyzed with a low-magnesium peritoneal dialysis solution. Perit Dial Int. 1995;15(1):61-64. 24. Brannan PG, Vergne-Marini P, Pak CY, Hull AR, Fordtran JS. Magnesium absorption in the human small intestine. Results in normal subjects, patients with chronic renal disease, and patients with absorptive hypercalciuria. J Clin Invest. 1976;57(6): 1412-1418. 25. Kwong MBL, Lee JSK, Chan MK. Transperitoneal calcium and magnesium transfer during an 8-hour dialysis. Perit Dial Int. 1987;7(2):85-89. 26. Spiegel DM. The role of magnesium binders in chronic kidney disease. Semin Dial. 2007;20(4):333-336. 27. Graf H, Kovarik J, Stummvoll HK, Wolf A. Disappearance of uraemic pruritus after lowering dialysate magnesium concentration. Br Med J. 1979;2(6203):1478-1479. 28. Gonella M, Ballanti P, Della Rocca C, et al. Improved bone morphology by normalizing serum magnesium in chronically hemodialyzed patients. Miner Electrolyte Metab. 1988;14(4): 240-245. 29. Gonella M, Buzzigoli G, Bencivelli W, Bartolini V, Betti G. The determination of whole blood magnesium concentration in uremics on chronic dialysis. Nephron. 1981;28(2):88-89. 30. Nilsson P, Johansson SG, Danielson BG. Magnesium studies in hemodialysis patients before and after treatment with low dialysate magnesium. Nephron. 1984;37(1):25-29. 31. Saha H, Harmoinen A, Pietilä K, Mörsky P, Pasternack A. Measurement of serum ionized versus total levels of magnesium and calcium in hemodialysis patients. Clin Nephrol. 1996;46(5): 326-331. 32. Navarro JF, Mora C, Jiménez A, Torres A, Macía M, García J. Relationship between serum magnesium and parathyroid hormone levels in hemodialysis patients. Am J Kidney Dis. 1999;34(1):43-48. 33. Blumenkrantz MJ, Kopple JD, Moran JK, Coburn JW. Metabolic balance studies and dietary protein requirements in patients undergoing continuous ambulatory peritoneal dialysis. Kidney Int. 1982;21(6):849-861. 34. Hutchison AJ, Were AJ, Boulton HF, Mawer EB, Laing I, Gokal R. Hypercalcaemia, hypermagnesaemia, hyperphosphataemia and hyperaluminaemia in CAPD: improvement in serum biochemistry by reduction in dialysate calcium and magnesium concentrations. Nephron. 1996;72(1):52-58. 35. Saha HH, Harmoinen AP, Pasternack AI. Measurement of serum ionized magnesium in CAPD patients. Perit Dial Int. 1997;17(4):347-352. 36. Katopodis KP, Koliousi EL, Andrikos EK, Pappas MV, Elisaf MS, Siamopoulos KC. Magnesium homeostasis in patients undergoing continuous ambulatory peritoneal dialysis: role of the dialysate magnesium concentration. Artif Organs. 2003;27:853-857. 37. Eisenman K, Holley JL. A higher magnesium dialysate concentration treats hypomagnesemia. Perit Dial Int. 2005;25(6): 604-605. 8 38. Ye H, Zhang X, Guo Q, et al. Prevalence and factors associated with hypomagnesemia in Southern Chinese continuous ambulatory peritoneal dialysis patients. Perit Dial Int. 2013;3(4): 450-454. 39. Wang S, McDonnell EH, Sedor FA, Toffaletti JG. pH effects on measurements of ionized calcium and ionized magnesium in blood. Arch Pathol Lab Med. 2002;126(8):947-950. 40. Walser M. Ion association. VI. Interactions between calcium, magnesium, inorganic phosphate, citrate, and protein in normal human plasma. J Clin Invest. 1961;40:723-730. 41. Takaya J, Higashino H, Kobayashi Y. Intracellular magnesium and insulin resistance. Magnes Res. 2004;17(2):126-136. 42. Romani AM, Matthews VD, Scarpa A. Parallel stimulation of glucose and Mg(21) accumulation by insulin in rat hearts and cardiac ventricular myocytes. Circ Res. 2000;86(3):326-333. 43. Kyriazis J, Kalogeropoulou K, Bilirakis L, et al. Dialysate magnesium level and blood pressure. Kidney Int. 2004;66(3): 1221-1231. 44. Elsharkawy MM, Youssef AM, Zayoon MY. Intradialytic changes of serum magnesium and their relation to hypotensive episodes in hemodialysis patients on different dialysates. Hemodial Int. 2006;10(suppl 2):S16-S23. 45. Pakfetrat M, Roozbeh Shahroodi J, Malekmakan L, Zare N, Hashemi Nasab M, Hossein Nikoo M. Is there an association between intradialytic hypotension and serum magnesium changes? Hemodial Int. 2010;14(4):492-497. 46. Loeb HS, Pietras RJ, Gunnar RM, Tobin JR Jr. Paroxysmal ventricular fibrillation in two patients with hypomagnesemia: treatment by transvenous pacing. Circulation. 1968;37:210215. 47. Ramee SR, White CJ, Svinarich JT, Watson TD, Fox RF. Torsade de pointes and magnesium deficiency. Am Heart J. 1985;109(1):164-167. 48. Beaubien ER, Pylypchuk GB, Akhtar J, Biem HJ. Value of corrected QT interval dispersion in identifying patients initiating dialysis at increased risk of total and cardiovascular mortality. Am J Kidney Dis. 2002;39(4):834-842. 49. Averbukh Z, Rosenberg R, Galperin E. Cell-associated magnesium and QT dispersion in hemodialysis patients. Am J Kidney Dis. 2003;41(1):196-202. 50. Randall RE Jr, Cohen MD, Spray CC Jr Rossmeisl EC. Hypermagnesemia in renal failure. Etiology and toxic manifestations. Ann Intern Med. 1964;61:73-88. 51. Cheng PT, Grabher JJ, LeGeros RZ. Effects of magnesium on calcium phosphate formation. Magnesium. 1988;7(3):123-132. 52. Kircelli F, Peter ME, Sevinc Ok E, et al. Magnesium reduces calcification in bovine vascular smooth muscle cells in a dose-dependent manner. Nephrol Dial Transplant. 2012;27(2): 514-521. 53. Montes de Oca A, Guerrero F, Martinez-Moreno JM, et al. Magnesium inhibits Wnt/b-catenin activity and reverses the osteogenic transformation of vascular smooth muscle cells. PLoS One. 2014;9(2):e89525. 54. Spiegel DM, Farmer B. Long-term effects of magnesium carbonate on coronary artery calcification and bone mineral density in hemodialysis patients: a pilot study. Hemodial Int. 2009;13(4):453-459. 55. Turgut F, Kanbay M, Metin MR, et al. Magnesium supplementation helps to improve carotid intima media thickness in patients on hemodialysis. Int Urol Nephrol. 2008;40(4): 1075-1082. 56. Tzanakis IP, Stamataki EE, Papadaki AN, Giannakis N, Damianakis NE, Oreopoulos DG. Magnesium retards the progress Am J Kidney Dis. 2015;-(-):--FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce Magnesium and Dialysis of the arterial calcifications in hemodialysis patients: a pilot study. Int Urol Nephrol. 2014;46(11):2199-2205. 57. Navarro JF, Mora C, Macia M, Garcia J. Serum magnesium concentration is an independent predictor of parathyroid hormone levels in peritoneal dialysis patients. Perit Dial Int. 1999;19(5):455-461. 58. Sánchez C1, López-Barea F, Sánchez-Cabezudo J, et al. Low vs standard calcium dialysate in peritoneal dialysis: differences in treatment, biochemistry and bone histomorphometry. A randomized multicenter study. Nephrol Dial Transplant. 2004;19(6): 1587-1593. 59. Wei M, Esbaei K, Bargman JM, Oreopoulos DG. Inverse correlation between serum magnesium and parathyroid hormone in peritoneal dialysis patients: a contributing factor to adynamic bone disease? Int Urol Nephrol. 2006;38(2):317-322. 60. Pletka P, Bernstein DS, Hampers CL, Merrill JP, Sherwood LM. Relationship between magnesium and secondary hyperparathyroidism during long-term hemodialysis. Metabolism. 1974;23(7):619-630. 61. Parsons V, Papapoulos SE, Weston MJ, Tomlinson S, O’Riordan JL. The long-term effect of lowering dialysate magnesium on circulating parathyroid hormone in patients on regular haemodialysis therapy. Acta Endocrinol (Copenh). 1980;93(4):455-460. 62. Ohya M1, Negi S, Sakaguchi T, et al. Significance of serum magnesium as an independent correlative factor on the parathyroid hormone level in uremic patients. J Clin Endocrinol Metab. 2014;99(10):3873-3878. 63. Iguchi A1, Watanabe Y, Iino N, Kazama JJ, Iesato H, Narita I. Serum magnesium concentration is inversely associated with fibroblast growth factor 23 in haemodialysis patients. Nephrology (Carlton). 2014;19(11):667-671. 64. Rodríguez-Ortiz ME, Canalejo A, Herencia C, et al. Magnesium modulates parathyroid hormone secretion and upregulates parathyroid receptor expression at moderately low calcium concentration. Nephrol Dial Transplant. 2014;29(2):282-289. 65. Rude RK, Singer FR, Gruber HE. Skeletal and hormonal effects of magnesium deficiency. J Am Coll Nutr. 2009;28(2): 131-141. 66. Saito N, Tabata N, Saito S. Bone mineral density, serum albumin and serum magnesium. J Am Coll Nutr. 2004;23(6):701S703S. 67. Markaki A, Kyriazis J, Stylianou K, et al. The role of serum magnesium and calcium on the association between adiponectin levels and all-cause mortality in end-stage renal disease patients. PLoS One. 2012;7:e52350. 68. Courivaud C, Davenport A. Magnesium and the risk of allcause and cardiac mortality in hemodialysis patients: agent provocateur or innocent bystander? Kidney Int. 2014;85(1):17-20. 69. Jahnen-Dechent W, Ketteler M. Magnesium basics. Clin Kidney J. 2012;5(suppl 1):i3-i14. 70. de Baaij JHF, Hoenderop JGJ, Bindels RJM. Regulation of magnesium balance: lessons learned from human genetic disease. Clin Kidney J. 2012;5(suppl 1):i15-i24. Am J Kidney Dis. 2015;-(-):--FLA 5.2.0 DTD Š YAJKD55344_proof Š 3 April 2015 Š 5:19 pm Š ce 9