Magnesium Research 2005; 18 (2): 123-6 ORIGINAL ARTICLE

Review paper

Magnesium therapy for nephrolithiasis

L. Massey
Food Science and Human Nutrition, Washington State University, PO Box 1495,
Spokane WA 99210-1495, USA
Correspondence: Linda Massey, Food Science and Human Nutrition, Washington State University, PO Box 1495,
Spokane WA 99210-1495, USA. <massey@wsu.edu>

Abstract. Purpose. Critically evaluate the experimental evidence and clinical trial
outcomes as the basis for use of magnesium (Mg) supplements as therapy for
calcium oxalate nephrolithiasis. Materials and methods. Literature search of
MedLine and Web of Science through January 2005; articles cited in papers found by
searches. Results. Magnesium inhibits calcium oxalate crystallization in human
urine and model systems. Magnesium also inhibits absorption of dietary oxalate
from the gut lumen. Three early trials of Mg oxide (MgO) and Mg hydroxide
(Mg(OH)2) reported lower rates of recurrent stone formation. However in a double-
blind, randomized, placebo-controlled trial with more carefully selected patients,
there was no significant difference between recurrence rates with 650 or 1300 mg
MgO daily and the placebo. Another trial reported 391 mg (21 meq) Mg daily as a
mixed salt, Mg potassium citrate, reduced calcium stone recurrence by 90%, similar
to potassium citrate, but with better gastrointestinal tolerance. The failure of MgO
and Mg(OH)2 as sole therapy may be related to poor absorption and low rates of Mg
deficiency in the patient populations tested. Conclusions. Clinical trial evidence
does not justify the use of MgO or Mg(OH)2 as a sole therapy for calcium oxalate
kidney stones in a general patient population. However, the addition of magnesium
to potassium citrate therapy improves outcomes. Clinical trials should focus on
patients who are likely to be Mg deficient.
Key words: magnesium, calcium, oxalate, nephrolithiasis

Mg acts as a competitor to calcium in oxalate bind-
ing. However, Mg oxalate (MgOx) is more soluble
than calcium oxalate (CaOx), 0.07 g/100 mL versus
0.0007 g/100 mL respectively, so MgOx does not form
stones at physiological urine concentrations.
Because Mg competes with Ca in binding oxalate,
both in the gut and urine, the ratio of Mg/Ca in the
urine has been used as an estimate of stone risk.
Because the absolute amount of oxalate to the total
amount of these divalent cations is also important,
the Mg/Ca ratio is less predictive of stone risk than
when other urinary factors are included in the calcu-
lation of calcium oxalate saturation. For example,
the relationship of Mg as an inhibitor of CaOx pre-
cipitability in 24 h urine is shown in the Tiselius Risk
Index (TRI) equation [1]. A TRI of 1.0 equals the
saturation level of CaOx :
0.84 -0.12 -0.22 -1.03
1.9×[Ox]×[Ca] ×[Mg] ×[Cit] ×[Volume]
In actual urine, the TRI may exceed 1.0 without crys-
tallization because it is in a metastable condition;
also, inhibitory proteins are found in the urine [2]. In
practice, Tiselius found the relative risk of stone
recurrence in 8 years doubled only when the TRI
exceeded 1.5 [3].
Another way of examining the effect of Mg on rela-
tive risk is to study crystallization of CaOx in model
systems. Rodgers et al. [4] found that Mg in combina-
tion with citrate, another inhibitor, is more effective
than either alone. Mg citrate slowed crystal growth
rate, nucleation rate and supersaturation in his
model system. A follow-up study [5] found Mg
supplements alone had no effect, but Mg with citrate
increased pH and lowered the relative saturation of
brushite in urine.
A second way Mg may reduce the CaOx stone risk is
through its effect on oxalate absorption. Even
though MgOx is 100 times more soluble than CaOx, it

123
L. MASSEY
is still relatively insoluble. Because of this property,
Mg binds oxalate in the gut and reduces its absorp-
tion. Liebman and Costa [6] loaded healthy volun-
teers with 198 mg oxalate containing 18 mg C13
oxalate. Simultaneous consumption of 300 mg either
elemental calcium as the carbonate salt or Mg as the
oxide salt reduced oxalate absorption from 13.5% to
5.1% and 7.6% respectively. These results confirmed
the 1986 report of Berg et al. [7] who used 500 g
spinach as the oxalate load, which contained
2027 mg oxalate, and Mg as alkaline Mg carbonate.
In rats, Mg deficiency causes nephrocalcinosis and
stone formation [8] and Mg supplements reduced
nephrolithiasis [9]. Thus, both theoretical consider-
ations and animal studies support trials of Mg supple-
ments as therapy for nephrolithiasis.
Methods
Literature search of MedLine and Web of Science
through January 2005; articles cited in papers found
on MedLine and Web of Science.
Results
In 1929 Hammarsten [10] showed that a deficiency of
magnesium led to an increase in the urinary excre-
tion of oxalate in humans, the process being reversed
when magnesium was supplied. The extensive early
studies on magnesium deficiency and the kidney
were reviewed by Thomas and Durlach [11] in 1971.
De Albuquerque and Tuma [12] reported the first
human study in nephrolithiasis patients in 1962;
which found that 3.7 mmol (150 mg) of Mg as MgO
three times daily significantly reduced oxalate excre-
tion and crystalluria in 9 persons with recurrent
oxalate lithiasis. The first clinical trial in which stone
recurrence was monitored was reported by Moore
and Bunce [13], who gave 10.4 mmol Mg (420 mg) as
MgO therapy to two patients with recurrent calcium
stones and found no recurrence in one year of
therapy. Melnick et al. [14] found in a larger trial with
33 patients that 17.9 mmol Mg (720 mg) as MgO
divided into two daily doses, reduced the pre-
treatment rate of 1.41 to 0.36 stones per year post-
treatment, a 75% reduction. No control group was
included. In a similar trial with a control group,
Johansson et al. [15] also found positive results of
4.4 mmol (400 mg) of Mg as Mg(OH)2 divided into
two daily doses. Patients with no treatment had a
44% recurrence rate over 2-3 years compared to 12%
in the Mg treatment group, a 73% reduction. How-
ever, the stone recurrence rate dropped in both
124
groups: from 1.25 to 0.03 stones per year in the treat-
ment group and from 0.5 to 0.22 in the no treatment
group. In contrast, in a double-blind, randomized,
placebo-controlled trial, Ettinger et al. [16] were not
able to show a difference between 16.1 mmol or
32.2 mmol Mg as MgO daily and the placebo recur-
rence rates in 51 patients with a history of at least
two stones. Recurrence rates over at least two years
were reduced by 56-65% in all groups.
Adding magnesium to potassium citrate appears to
change urine composition more favorably than using
potassium citrate alone. Pak et al. [17] found that not
only was urinary magnesium higher, but also pH and
citrate. Although urinary oxalate was reduced from
829 to 716 lmol/d (about the 7% reduction expected),
this decrease was not statistically significant, but
contributed to the decrease in calcium oxalate activ-
ity and formation products. Kato et al. [18] reported
similar findings using two supplements, potassium-
sodium citrate plus 6.2 mmol (250 mg) MgO.
Recently Ettinger et al. [19] confirmed the effective-
ness of the use of magnesium potassium citrate in a
randomized clinical trial. They found 12.7 mmol Mg
(510 mg) daily of Mg potassium citrate reduced cal-
cium stone recurrence over 30 months by 90%, simi-
lar to potassium citrate, but with better gastrointesti-
nal tolerance.
Discussion
The failure of MgOx and Mg(OH)2 as sole therapy
may be related to two factors. First is their poor
absorption. MgO and Mg(OH)2 are the least bioavail-
able Mg salts [20]. The more soluble forms of Mg are
chloride, gluconate, aspartate and citrate. Mg citrate
doubled the Mg/Cr ratio 2-4 hours after oral loading,
while MgO only increased the ratio by 3% [21].
Second, Mg deficiency is relatively rare in stone
formers. Schwartz et al. [22] reported that only 11%
of 24 h urinary Mg in stoneformers falls below
43 mg/d. Preminger et al. [23] reported that 4.3% of a
series of 1116 patients had hypomagnesiuria, which
they defined as having a urinary magnesium level of
less than 50 mg/d. A large clinical laboratory
(Litholink, Chicago IL) uses 44 mg/d as the lower
limit for urinary Mg based on the 5th percentile of
their normal population. In stoneformers over 18 y,
1434 of 31,300 urine collections (4.6%) fell below this
cut point (J. Asplin, personal communication, Janu-
ary 6, 2004).
Even if a Mg salt is well absorbed and increases
urinary Mg, most patients already have sufficient
urinary Mg to decrease crystallization to its mini-

Figure 1. Changes in Tiselius Risk Index for calcium
oxalate saturation with changes in urinary magnesiu-
m. Curve A UCa = 6.39 mmol/d, UOx = 600 lmol/d.
Curve B UCa = 8.0 mmol/d, UOx = 409 lmol/d.
Curve C UCa = 6.39 mmol/d, UOx = 409 lmol/d. The
horizontal dotted line is set at a TRI of 1.5, above
which symptomatic kidney stone risk doubles.
mum potential. In figure 1, the effect of increasing
Mg excretion on TRI is seen. For this figure, the
assumption for curve A is that urinary calcium is
constant at 6.39 mmol/d, citrate at 2.9 mmol/d,
oxalate at 409 lmol/d and volume at 1.86 mL, the
average urinary composition of stoneformers as
reported by Ettinger et al. [18]. However, increasing
daily Mg excretion from 1.5 mmol (37 mg) to
4.0 mmol (97 mg) may result in a clinically signifi-
cant decrease in risk if a TRI cut-off of 1.5 is used as
the critical value (dotted line); this level of change
reduces the TRI from 1.54 to 1.30. From Fig. 1, it can
be seen that a similar conclusion can be made for
hyperoxaluric patients (curve A) and hypercalciuric
patients (curve B). Overall, magnesium-deficient
patients who excrete less than 50 mg/day are most
likely to benefit from Mg supplements.
However, even a relatively insoluble Mg supple-
ment will reduce oxalate absorption even if there is
little increase in Mg excretion. As an example, if a Mg
supplement reduced dietary Ox absorption by 6%
with a daily consumption of 200 mg Ox as reported
by Liebman and Costa [6], a 125 lmol/d (12 mg) drop,
MAGNESIUM AND NEPHROLITHIASIS
the TRI would decrease from 1.5 to 1.1 with Mg
assumed to be constant at 5 mmol/d (121 mg) and
other conditions as described above. Because both
effects are fairly small, only a Mg supplement that
both binds oxalate in the gut and increases urinary
Mg excretion is likely to reduce the risk of stone
recurrence. For example, if Mg supplementation
both increased urinary Mg from 1.5 to 4.5 mmol/d
and reduced oxalate from 475 to 350 lmoles/d, the
TRI would decrease from 1.7 to 1.1.
Epidemiological evidence supports the probable
effectiveness of increased magnesium in preventing
symptomatic kidney stones. Taylor et al. [24] fol-
lowed 45,619 men for 14 years, with dietary assess-
ments every four years. After multivariate adjust-
ment the relative risk for the highest quintile of
magnesium intake (over 450 mg/d) compared to the
lowest quintile (less than 314 mg/d) was 0.71.
Clinical trials of Mg supplements should include
assessment of the Mg status of trial participants by
urinary Mg/d, serum Mg and dietary Mg intakes.
Dietary Mg is unlikely to be deficient if the diet
includes green leafy vegetables and whole grains, as
Mg is a major mineral in chlorophyll. Chocolate and
nuts are good dietary sources of Mg, but also contain
considerable amounts of oxalate which increases
urinary oxalate [25, 26]. The trial design should
require that the Mg supplement be taken with meals
to maximize oxalate binding. Inclusion of both Mg
deficient and Mg sufficient groups would help clarify
which nephrolithiasis patients could benefit from Mg
supplements.
References
1. Tiselius HG. Aspects on estimation of the risk of calcium
oxalate crystallization in urine. Urol Res 1991 ; 47 : 255-9.
2. Borghi L, Meschi T, Guerra A, Beramaschi E, Mutti A,
Novarini A. Effects of urinary macromolecules on the
nucleation of calcium oxalate in idiopathic stone formers
and healthy controls. Clin Chim Acta 1995 ; 239 : 1-11.
3. Tiselius HG. Factors influencing the course of calcium
oxalate stone disease. Eur Urol 1999 ; 36 : 363-70.
4. Rodgers A. Aspects of calcium oxalate crystallization :
Theory, in vitro studies and in vivo implementation. J
Am Soc Nephrol 1999 ; 10 : S351-S354.
5. Allie S, Rodgers A. Effects of calcium carbonate, magne-
sium oxide and sodium citrate bicarbonate health supple-
ments on the urinary risk factors for kidney stone forma-
tion. Clin Chem Lab Med 2003 ; 41 : 39-45.
6. Liebman M, Costa G. Effects of calcium and magnesium
on urinary oxalate excretion after oxalate loads. J Urol
2000 ; 163 : 1565-9.
125
L. MASSEY
7. Berg W, Bothor C, Pirlich W, Janitzky V. Influence of mag-
nesium on the absorption and excretion of calcium and
oxalate ions. Eur Urol 1986 ; 12 : 274-82.
8. Ko KW, Fellers FX, Craig JM. Observations on magne-
sium deficiency in the rat. Lab Inv 1962 ; 11 : 294-305.
9. Khan SR, Shevock PN, Hackett RL. Magnesium oxide
administration and prevention of calcium oxalate neph-
rolithiasis. J Urol 1993 ; 149 : 412-6.
10. Hammarsten G. On calcium oxalate and its solubility in
the presence of inorganic salts with special reference to
occurrence of crystaluria. C R Trav Lab Carlsberg 1929 ;
17 : 1-85.
11. Thomas T, Durlach J. Déficit magnésique et pathologie
du rein. In : 1er Symposium International sur le déficit
magnésique en pathologie humaine. 1. Volume des rap-
ports. J Durlach, ed., 359–85.
12. De Albuquerque PF, Tuma M. Investigations on urolithi-
asis. II. Studies on oxalate. J Urol 1962 ; 87 : 504-6.
13. Moore CA, Bunce GE. Reduction in frequency of renal
calculus formation by oral magnesium administration.
A preliminary report. Inv Urol 1964 ; 2 : 7-13.
14. Melnick I, Landes RR, Hoffman AA, Burch JF. Magne-
sium therapy for recurring calcium oxalate urinary cal-
culi. J Urol 1971 ; 105 : 119-22.
15. Johansson G, Backman U, Danielson BG, Fellstrom B,
Ljunghall S, Wikstrom B. Biochemical and clinical
effects of the prophylactic treatment of renal calcium
stones with magnesium hydroxide. J Urol 1980 ; 124 :
770-4.
16. Ettinger B, Citron JT, Livermoore B, Dolman LI. Chlo-
rthalidone reduces calcium oxalate calculous recur-
rence but magnesium hydroxide does not. J Urol 1988 ;
139 : 679-84.
126
17. Pak CYC, Koenig K, Khan R, Haynes S, Padalino P.
Physicochemical action of potassium-magnesium cit-
rate in nephrolithiasis. J Bone Min Res 1992 ; 7 : 281-5.
18. Kato Y, Yamaguchi S, Yachiku S, Nakazono S, Hori JI,
Wada N, Hou K. Changes in urinary parameters after
oral administration of potassium-sodium citrate and
magnesium oxide to prevent urolithiasis. Urol 2004 ; 63 :
7-12.
19. Ettinger B, Pak CYC, Citron JT, Thomas C, Adams-
Huet B, Vangessel A. Potassium-magnesium citrate is an
effective prophylaxis against recurrent calcium oxalate
nephrolithiasis. J Urol 1997 ; 158 : 2069-73.
20. Lindberg JS, Zobitz MM, Poindexter JR, Pak CYC. Mag-
nesium bioavailability from magnesium citrate and mag-
nesium oxide. J Am Coll Nutr 1990 ; 9 : 48-55.
21. Ranade VV, Somberg JC. Bioavailability and pharmaco-
kinetics of magnesium after administration of magne-
sium salts to humans. Am J Ther 2001 ; 8 : 345-57.
22. Schwartz BF, Bruce J, Leslie S, Stoller ML. Rethinking
the role of urinary magnesium in calcium urolithiasis. J
Endourol 2001 ; 15 : 233-5.
23. Preminger GM, Baker S, Peterson R, Poindexter J,
Pak CYC. Hypomagnesiuric hypocitraturia : An appar-
ent new entity for calcium nephrolithiasis. J Lithot-
rypsy Stone Disease 1989 ; 1 : 22-5.
24. Taylor EN, Stampfer MJ, Curhan GC. Dietary factors
and the risk of incident kidney stones in men : New
insights after 14 years of follow-up. J Am Soc Nephrol
2004 ; 15 : 3225-32.
25. Chai W, Liebman M. Oxalate content of legumes, nuts
and grain-based flours. J Food Composition Analysis
2005 ; (in press).
26. Mendonca C, Martini LA, Baxmann AC, et al. Effects of
an oxalate load on urinary oxalate excretion in calcium
stone formers. J Ren Nutr 2003 ; 13 : 39-46.