British Journal of Urology (1975), 47, 237-242
The Relationship of the Urinary Cations, Calcium, Magnesium, Sodium and Potassium, in Patients with Renal Calculi s.
G . WELSHMAN
and
M.
c. MCCEOWN
Belfast City Hospital, Belfast
One of the main points of interest in the study of renal calculi formation has been the quantitative examination of urinary constituents. The high incidence of calcium-containing stones in hyperparathyroid and hypercalciuric patients seems closely related to an increased excretion of urinary calcium. However, it is also well established that a considerable number of people with high urinary excretion of calcium do not form stones, and that renal calculi often occur in patients who are normo-calciuric. In recent years, interest has been shown in the excretion of other urinary cations, particularly magnesium and sodium, which may increase the solubility of calcium crystals by displacing the calcium ions from the crystal lattice. Evans et al. (1967) found magnesium to be reduced in the 24-hour urines of stone-formers with a normal calcium excretion, and Oreopoulos, Soyannwo and McGeown (1968) reported significantly reduced magnesium/calcium ratios in the early-morning urines of fasting stone-forming patients. Modlin (1967) studied the excretion of sodium in stone-formers, whites and Bantu. He observed that white stone-formers exhibited both a reduced 24-hour sodium excretion and sodium/calcium ratio compared with normal white subjects, whereas the Bantu, who had a very low incidence of renal calculi, exhibited a very high sodium excretion. Controversy has arisen concerning the results of these investigations and also with regard to the use of ratios to compare the excretion of any two urinary constituents. In this study, the mean 24-hour urinary excretions of calcium, magnesium, sodium and potassium was estimated in a large number of confirmed stone-formers in whom hyperparathyroidism could be fairly confidently excluded and the results compared with those of normal young adults and a large group of patients free of any renal disorders. The excretion of each constituent relative to calcium was determined by calculating the coefficient of correlation and the linear regression equation. As stone-forming crystals may be deposited from highly concentrated urine samples, calcium, magnesium, sodium and potassium concentrations relative to creatinine were also determined in early-morning urine specimens of stone-formers and controls. Only male subjects in the patient and control groups were used in this investigation as there is a marked sex difference in the excretion of calcium ions (Bulusu et al., 1970; Davis, Morgan and Rivlin, 1970). Material and Methods 24-hour urine collections were obtained from 101 male patients with calcium-stone disease admitted to the Belfast City Hospital over a 3-year period. The presence of this type of stone was confirmed by chemical analysis or by the presence of a dense radio-opaque stone in the renal tract. Patients with possible parathyroid disorders were excluded from this investigation, as were cases with renal tubular acidosis, sarcoidosis and vitamin D intoxication. As renal malfunction affects the urinary excretion of cations, all subjects with blood urea greater than 45 mg/100 ml 47/3-~
237
238
BRITISH JOURNAL OF UROLOGY
were also excluded. 24-hour urine collections sent to the laboratory from 89 male patients with various non-renal diseases were used as patient controls. It was considered that results from alarge number of patients might provide a more appropriate normal range for comparison with hospitalised stone-formers. A second set of normal controls was obtained from urine collections of 63 young male adults (aged 17 to 30 years), consisting of army personnel and laboratory workers. All subjects were on a free diet, except that stone-formers were encouraged to maintain a high fluid intake. It was assumed that individual variations in diet would be masked by the relatively large number of persons, from a homogeneous population, used in this investigation. The difference in diet at home and in hospital should be minimal, as every attempt is made by this hospital to provide a diet similar to that consumed by the patients at home. Fresh early-morning specimens were obtained on wakening, from 43 normal male adults and from 41 male stone-formers. All subjects were selected by the same criteria used for the 24-hour collections. Calcium was determined by an autoanalysertechnique (Technicon Methodology N-3b), involving the reaction of cresolphthalein complexone and diethylamine with calcium. 8-hydroxyquinoline was added to the reaction to suppress interference by magnesium. Magnesium was estimated by the colorimetric reaction with Mann Yoe’s azo dye, using a test kit produced by Roche Diagnostics. All calcium and magnesium results were rechecked by an atomic absorption spectrophotometricmethod. The tests were repeated if there was any significant difference in the results of the two methods. Sodium and potassium were analysed by flame emission spectrophotometry.Creatinine was determined by the alkaline picrate reaction, using an autoanalyser (Technicon Methodology N-1 1 b).
Results
The mean 24-hour urinary excretions of calcium, magnesium, sodium and potassium were estimated for stone-formers, stone-free patients and young normal adults (Table I). There was no significant difference in the mean calcium, magnesium and sodium excretions between stoneformers and patient controls (P> 0.05). Young adult controls showed no significant difference in calcium and magnesium excretion (P> 0.05) when compared with stone-formers, but there is a highly significant increase in their sodium excretion (P< 0.001). The mean urinary potassium excretion for stone-formers is significantly lower (P< 0.001) than the values obtained from both control groups. When the results are expressed in relation to I g of urinary creatinine (Table 11), the comparative excretion pattern of the 4 cations, between stone-formers and controls, remains unchanged, except that the stone-formers had a moderately significant increase in calcium compared with male patient controls (P< 0.05). A comparison of the 24-hour urinary excretions between calcium and magnesium, calcium and sodium, and calcium and potassium were determined by calculating the correlation coefficient and linear regression equations. The results for the 3 groups are shown in Table 111. Significant correlations between the cation pairs for all groups were observed. Analysis of covariance was used to compare the linear regression lines of the stone-formers, patient controls and young adult controls, for each pair of cations. The regression lines for calcium on magnesium did not differ in either slope or position (P>0*05). There was also no significant difference in the 3 regression lines for calcium on sodium, for in either slope or position (P > 0.05). The relative calcium and potassium excretions of stone-formers and patient controls were not significantly different (P>0.05)for slope and position. However, the regression line for the young adult controls had a significantly different slope (P 0.05). The urines of stone-formers showed a significantly higher mean concentration
RELATIONSHIP OF URINARY CATIONS IN PATIENTS WITH RENAL CALCULI
239
Table I The mean 24-hour Urinary Excretions ( f S.E.M.) of Calcium, Magnesium, Sodium and Potassium in Male Stone-formers, Patient Controls and Young Adult Controls Group
No.
Stone-formers
101
Patient controls
89
Young adult controls
63
Calcium mg/24 hr
Magnesium mg/24 hr
204 (8.7) 183 (8.9) 21 3 (9.1)
83 (3.5) 87 (3.7) 90 (3.9)
Sodium mEq/24 hr
Potassium mEq/24 hr
Table 11 The Mean Concentration per gram of Creatinine ( f S.E.M.) of Calcium, Magnesium, Sodium and Potassium in 24-hour Urine Collection from Male Stone-formers, Patient Controls and Young Adult Controls Calcium mg/l g Creat.
Magnesium mg/l g Creat.
Sodium mEq/l g Creat.
Potassium mEq/l g Creat.
101
158 (6.9)
138 (6.7) 146 (9.3)
104 (4.9) 104 (5.2) 127 (5.9)
32 (1.3)
89
66 (2.4) 40 (2.9)
Group
No.
Stone-formers Patient controls Young adult controls
59
64 (2.9)
44 (1.8) 47 (2.0)
Table III The Linear Regression Equations and Correlation Coefficients for Calcium on Magnesium, Calcium on Sodium and Calcium on Potassium Linear Regression Equations
Correlation Coefficient
Probability t Test
Stone-formers Patient controls Young adult controls
Magnesium = 44.2+0.19 calcium Magnesium = 62.8 0.13 calcium Magnesium = 51*4+0.18calcium
r = 0.494
r = 0442 r = 0.556
P < 0.001 P < 0.001 P < 0.001
S tone-formers Patient controls Young adult controls
Sodium Sodium Sodium
r = 0.341 r = 0.305 r = 0.471
P0.05) in slope or position between the calcium on magnesium regression lines of stone-formers (Mg = 27.2 k 0.26 Ca) and controls (Mg = 49.0 5 0.31 Ca).
Discussion Normal values for calcium, magnesium, sodium and potassium were obtained by analysing the 24-hour urine collections from young healthy males. As this group may not give a true normal comparison with hospitalised stone-formers, for reasons of differences in age, activity and physical fitness, a second set of controls was acquired from the urines of a large number of hospital in-patients known to be free of any renal abnormalities. The mean daily excretion of urinary calcium for the normal controls was similar to the values obtained by Sutton and Watson (1969) and Bulusu et al. (1970). The stone-formers, however, did not differ significantly in the daily calcium excretion from these normal controls. This is contrary to the observations of Evans et al. (1967), Bulusu eta/. (1970) and White et a/. (1969), who found the mean daily calcium excretion for male idiopathic stone-formers to be 380 mg, 338 mg and 320 mg respectively. However, King et al. (1968) obtained values of only 21 1 mg per 24 hours and a mean value of 200 mg for young adult male controls. According to King (1971) only about 10% of stone-formers excreted more than 300 mg of calcium per day. In the present series the percentage of hypercalciurias (urinary calcium > 300 mg per 24 hours) in male stone-former, patient control and young adult control groups were 11.8, 10.2 and 10.8 respectively. The normal 24-hour magnesium excretions were similar to those given by Sutton and Watson (1969), while Evans et al. (1967) and Modlin (1967) obtained higher values for urinary magnesium
RELATIONSHIP OF URINARY CATIONS IN PATIENTS WITH RENAL CALCULI
24 1
in normal controls. According to the latter, there was no difference in the magnesium excretion between stone-formers and normal controls, but they found that stone-formers with normal calcium excretion showed a significant reduction in urinary magnesium. The results in this investigation suggest that no significant difference in magnesium excretion exists between any of the groups, in spite of the fact that most of the stone-formers are normocalciuric. With regard to the concentrations of calcium and magnesium, relative to creatinine, in early-morning urine, a moderately significant increase in calcium, and reduction in magnesium were observed for stoneformers compared with controls. These findings may be of some significance, as it is probable that the concentrated urine specimens are responsible for the deposition of stone-forming material. It has been reported by King et al. (1968) that the majority of stone-formers have an abnormally high calcium/magnesium ratio, calculated from 24-hour excretion values. When Oreopoulous el al. (1968) estimated the calcium/magnesium ratios from the calcium and magnesium concentrations in early-morning urine of stone-formers and controls, a significant increase in the mean calcium/magnesium ratio was observed in the stone-former's urine. The validity of ratios to relate 2 constituents and to compare these ratios between different groups have been criticised by Papadimitriou and Ram (1968) and Sutton and Watson (1968). In this study, the correlation coefficient and linear regression equation for 24-hour calcium on magnesium in each group were determined. The correlation was found to be highly significant in all groups, but there was no significant difference in the linear regression lines for calcium on magnesium between stoneformers and controls. With regard to the examination of early-morning urines, there was no significant difference in the calcium on magnesium linear regression equations for stone-formers and controls, in spite of the differences in calcium and magnesium concentrations. These results confirm the view that the use of ratios to compare excretory patterns between groups may be misleading. Modlin (1967) estimated sodium and calcium in white controls, Bantu and stone-formers. The Bantu, who are relatively free of stone disease, were reported to have a significantly higher excretion of sodium than normal white subjects, while stone-formers had a much lower excretion of sodium than normal controls. It was also stated that the mean sodium/calcium ratio was significantly lower in stone-formers than in white controls, and that the sodium/calcium ratio for the Bantu was much higher than the other two groups. On this evidence it was postulated that a high level of urinary sodium, and particularly an elevated sodium/calcium ratio, may prevent the occurrence of renal calculi. However, it has been found in this survey that stone-formers and normal controls excreted similar amounts of sodium per day. The young adult controls are exceptional in the fact that they excrete a greater amount of sodium, paralleling their increased excretion of other cations. The disparity between results for male patient controls (average age 47 years) and young healthy adults (average age 22 years) may be the reason for differences in interpretation of other workers. The relationships of calcium to sodium, expressed by the linear regression equations, were not significantly different. The urinary potassium excretion of male stone-formers was unexpectedly low in comparison with the values for both normal patients and young adults. There was no significant difference in the linear regression equations of stone-formers and patient controls, although it appears that the former have a lower excretion of potassium relative to calcium. There was no evidence to suggest that the potassium levels in the diet of stone-formers differs from that of the normal population. At present, no obvious explanation can be found for the low 24-hour potassium excretion of stone-formers. The early-morning urines of stone-formers and young adult controls have similar concentrations of sodium and potassium relative to creatinine.
Conclusion The 24-hour urinary excretions of calcium, magnesium and sodium by male idiopathic stoneformers do not appear to be significantly different from normal male controls. Potassium
242
BRITISH JOURNAL OF UROLOGY
excretion by stone-formers, however, was significantly lower than the controls. A moderately significant increase in calcium and a reduction in magnesium, relative to 1 g of creatinine, were observed in early-morning urines of patients with renal calculi. These abnormalities in concentrated urine samples which pass slowly through the renal tract overnight may be an important factor in the initiation and development of calcium-containing stones. Summary
Calcium, magnesium, sodium and potassium were estimated in the 24-hour urine collections of 101 idiopathic male stone-formers, 89 male patient controls and 59 young male adult controls. The results were calculated in terms of 24-hour volume and 1 g of creatinine. The concentrations of the 4 cations, relative to a gram of creatinine, were also determined in the early-morning urines of 41 male stone-formers and 43 young male adult controls. No difference was observed in the 24-hour excretion of calcium, magnesium and sodium between the stone-formers and controls. The mean daily potassium excretion, however, was significantly reduced in the urine of stone-formers. The linear regression equations were calculated for calcium on magnesium, calcium on sodium, and calcium on potassium, using the 24-hour excretion values of these cations. Only the calcium on potassium line of stone-formers was significantly different from that of the normal subjects. A significant increase by stone-formers in the urinary calcium concentration of their earlymorning specimens was found. The high concentration of urinary calcium in overnight urines of stone-formers combined with a low magnesium concentration might possibly contribute to the development of renal stone disease. The authors thank Dr J. D. Merrett, Department of Medical Statistics, Queen’s University, Belfast, for his help with the statistical analysis. They also wish to express their gratitude to Sister E. Cox and Sister I. Young for arranging the collection of specimens from their patients.
References BULUSU,L., HODGKINSON, A., NORDIN,B. E. C. and PEACOCK, M. (1970). Urinary excretion of calcium and creatinine in relation to age and body weight in normal subjects and patients with renal calculus. Clinical Science, 38, 601-612.
DAVIS,R. H., MORGAN, D. B. and RIVLIN,R. S. (1970). The excretion of calcium in the urine and its relation to calcium intake, sex and age. Clinical Science, 39, 1-12. EVANS,R. A., FORBES, M. A., SUTTON,R. A. L. and WATSON, L. (1967). Urinary excretion of calcium and magnesium in patients with calcium-containing renal stones.Lancet, 2,958-961. KINO,J. S. (1971). Currents in renal stone research. Clinical Chemistry, 17, 971-982. KING,J. S., O’CONNOR, F. J., SMITH,M. J. V. and CRouse, L. (1968). The urinary calcium/magnesium ratio in calcigerous stone formers. InvestigPtive Urology, 6, 60-65. MODLIN,M. (1967). The aetiology of renal stone: a new concept arising from studies on a stone-free population. Annals of the Royal College of Surgeons, 40,155-178. OREOPOULOS, D. G., SOYANNWO, M. A. and MCGEOWN,M. G. (1968). Magnesium/calcium ratio in urine of patients with renal stones. Lancet, 2,420-422. PAPADIMITRIOU, M. and RAM, M. D. (1968). Urinary magnesium and renal stones. Lancet, 2, 577. SUITON. R. A. L. and WATSON,L. (1969). Urinary excretion of calcium and magnesium in primary hyperparathyroidism. Lancet, i, 1000-1003. -(1968). Urinary magnesium and renal stones. Lancet, 2,636. WHITE,R. W., COHEN,R. D., VINCE,F. P., WILLIAMS, G., BLANDY, J. and TRESIDDER, G. C. (1969). Minerals in the urine of stone-formers and their spouses. In: Renal Stone Research Symposium, Ed. A. Hodgkinson and B. E. C. Nordin, London: Churchill, pp. 289-296.
The Authors S. G. Welshman, MSc, FRIC, Chief Clinical Biochemist. M. G. McGeown, MD, PhD,FRCPEd, Consultant Nephrologist.
The Relationship of the Urinary Cations, Calcium, Magnesium, Sodium and Potassium, in Patients with Renal Calculi s.
G . WELSHMAN
and
M.
c. MCCEOWN
Belfast City Hospital, Belfast
One of the main points of interest in the study of renal calculi formation has been the quantitative examination of urinary constituents. The high incidence of calcium-containing stones in hyperparathyroid and hypercalciuric patients seems closely related to an increased excretion of urinary calcium. However, it is also well established that a considerable number of people with high urinary excretion of calcium do not form stones, and that renal calculi often occur in patients who are normo-calciuric. In recent years, interest has been shown in the excretion of other urinary cations, particularly magnesium and sodium, which may increase the solubility of calcium crystals by displacing the calcium ions from the crystal lattice. Evans et al. (1967) found magnesium to be reduced in the 24-hour urines of stone-formers with a normal calcium excretion, and Oreopoulos, Soyannwo and McGeown (1968) reported significantly reduced magnesium/calcium ratios in the early-morning urines of fasting stone-forming patients. Modlin (1967) studied the excretion of sodium in stone-formers, whites and Bantu. He observed that white stone-formers exhibited both a reduced 24-hour sodium excretion and sodium/calcium ratio compared with normal white subjects, whereas the Bantu, who had a very low incidence of renal calculi, exhibited a very high sodium excretion. Controversy has arisen concerning the results of these investigations and also with regard to the use of ratios to compare the excretion of any two urinary constituents. In this study, the mean 24-hour urinary excretions of calcium, magnesium, sodium and potassium was estimated in a large number of confirmed stone-formers in whom hyperparathyroidism could be fairly confidently excluded and the results compared with those of normal young adults and a large group of patients free of any renal disorders. The excretion of each constituent relative to calcium was determined by calculating the coefficient of correlation and the linear regression equation. As stone-forming crystals may be deposited from highly concentrated urine samples, calcium, magnesium, sodium and potassium concentrations relative to creatinine were also determined in early-morning urine specimens of stone-formers and controls. Only male subjects in the patient and control groups were used in this investigation as there is a marked sex difference in the excretion of calcium ions (Bulusu et al., 1970; Davis, Morgan and Rivlin, 1970). Material and Methods 24-hour urine collections were obtained from 101 male patients with calcium-stone disease admitted to the Belfast City Hospital over a 3-year period. The presence of this type of stone was confirmed by chemical analysis or by the presence of a dense radio-opaque stone in the renal tract. Patients with possible parathyroid disorders were excluded from this investigation, as were cases with renal tubular acidosis, sarcoidosis and vitamin D intoxication. As renal malfunction affects the urinary excretion of cations, all subjects with blood urea greater than 45 mg/100 ml 47/3-~
237
238
BRITISH JOURNAL OF UROLOGY
were also excluded. 24-hour urine collections sent to the laboratory from 89 male patients with various non-renal diseases were used as patient controls. It was considered that results from alarge number of patients might provide a more appropriate normal range for comparison with hospitalised stone-formers. A second set of normal controls was obtained from urine collections of 63 young male adults (aged 17 to 30 years), consisting of army personnel and laboratory workers. All subjects were on a free diet, except that stone-formers were encouraged to maintain a high fluid intake. It was assumed that individual variations in diet would be masked by the relatively large number of persons, from a homogeneous population, used in this investigation. The difference in diet at home and in hospital should be minimal, as every attempt is made by this hospital to provide a diet similar to that consumed by the patients at home. Fresh early-morning specimens were obtained on wakening, from 43 normal male adults and from 41 male stone-formers. All subjects were selected by the same criteria used for the 24-hour collections. Calcium was determined by an autoanalysertechnique (Technicon Methodology N-3b), involving the reaction of cresolphthalein complexone and diethylamine with calcium. 8-hydroxyquinoline was added to the reaction to suppress interference by magnesium. Magnesium was estimated by the colorimetric reaction with Mann Yoe’s azo dye, using a test kit produced by Roche Diagnostics. All calcium and magnesium results were rechecked by an atomic absorption spectrophotometricmethod. The tests were repeated if there was any significant difference in the results of the two methods. Sodium and potassium were analysed by flame emission spectrophotometry.Creatinine was determined by the alkaline picrate reaction, using an autoanalyser (Technicon Methodology N-1 1 b).
Results
The mean 24-hour urinary excretions of calcium, magnesium, sodium and potassium were estimated for stone-formers, stone-free patients and young normal adults (Table I). There was no significant difference in the mean calcium, magnesium and sodium excretions between stoneformers and patient controls (P> 0.05). Young adult controls showed no significant difference in calcium and magnesium excretion (P> 0.05) when compared with stone-formers, but there is a highly significant increase in their sodium excretion (P< 0.001). The mean urinary potassium excretion for stone-formers is significantly lower (P< 0.001) than the values obtained from both control groups. When the results are expressed in relation to I g of urinary creatinine (Table 11), the comparative excretion pattern of the 4 cations, between stone-formers and controls, remains unchanged, except that the stone-formers had a moderately significant increase in calcium compared with male patient controls (P< 0.05). A comparison of the 24-hour urinary excretions between calcium and magnesium, calcium and sodium, and calcium and potassium were determined by calculating the correlation coefficient and linear regression equations. The results for the 3 groups are shown in Table 111. Significant correlations between the cation pairs for all groups were observed. Analysis of covariance was used to compare the linear regression lines of the stone-formers, patient controls and young adult controls, for each pair of cations. The regression lines for calcium on magnesium did not differ in either slope or position (P>0*05). There was also no significant difference in the 3 regression lines for calcium on sodium, for in either slope or position (P > 0.05). The relative calcium and potassium excretions of stone-formers and patient controls were not significantly different (P>0.05)for slope and position. However, the regression line for the young adult controls had a significantly different slope (P 0.05). The urines of stone-formers showed a significantly higher mean concentration
RELATIONSHIP OF URINARY CATIONS IN PATIENTS WITH RENAL CALCULI
239
Table I The mean 24-hour Urinary Excretions ( f S.E.M.) of Calcium, Magnesium, Sodium and Potassium in Male Stone-formers, Patient Controls and Young Adult Controls Group
No.
Stone-formers
101
Patient controls
89
Young adult controls
63
Calcium mg/24 hr
Magnesium mg/24 hr
204 (8.7) 183 (8.9) 21 3 (9.1)
83 (3.5) 87 (3.7) 90 (3.9)
Sodium mEq/24 hr
Potassium mEq/24 hr
Table 11 The Mean Concentration per gram of Creatinine ( f S.E.M.) of Calcium, Magnesium, Sodium and Potassium in 24-hour Urine Collection from Male Stone-formers, Patient Controls and Young Adult Controls Calcium mg/l g Creat.
Magnesium mg/l g Creat.
Sodium mEq/l g Creat.
Potassium mEq/l g Creat.
101
158 (6.9)
138 (6.7) 146 (9.3)
104 (4.9) 104 (5.2) 127 (5.9)
32 (1.3)
89
66 (2.4) 40 (2.9)
Group
No.
Stone-formers Patient controls Young adult controls
59
64 (2.9)
44 (1.8) 47 (2.0)
Table III The Linear Regression Equations and Correlation Coefficients for Calcium on Magnesium, Calcium on Sodium and Calcium on Potassium Linear Regression Equations
Correlation Coefficient
Probability t Test
Stone-formers Patient controls Young adult controls
Magnesium = 44.2+0.19 calcium Magnesium = 62.8 0.13 calcium Magnesium = 51*4+0.18calcium
r = 0.494
r = 0442 r = 0.556
P < 0.001 P < 0.001 P < 0.001
S tone-formers Patient controls Young adult controls
Sodium Sodium Sodium
r = 0.341 r = 0.305 r = 0.471
P0.05) in slope or position between the calcium on magnesium regression lines of stone-formers (Mg = 27.2 k 0.26 Ca) and controls (Mg = 49.0 5 0.31 Ca).
Discussion Normal values for calcium, magnesium, sodium and potassium were obtained by analysing the 24-hour urine collections from young healthy males. As this group may not give a true normal comparison with hospitalised stone-formers, for reasons of differences in age, activity and physical fitness, a second set of controls was acquired from the urines of a large number of hospital in-patients known to be free of any renal abnormalities. The mean daily excretion of urinary calcium for the normal controls was similar to the values obtained by Sutton and Watson (1969) and Bulusu et al. (1970). The stone-formers, however, did not differ significantly in the daily calcium excretion from these normal controls. This is contrary to the observations of Evans et al. (1967), Bulusu eta/. (1970) and White et a/. (1969), who found the mean daily calcium excretion for male idiopathic stone-formers to be 380 mg, 338 mg and 320 mg respectively. However, King et al. (1968) obtained values of only 21 1 mg per 24 hours and a mean value of 200 mg for young adult male controls. According to King (1971) only about 10% of stone-formers excreted more than 300 mg of calcium per day. In the present series the percentage of hypercalciurias (urinary calcium > 300 mg per 24 hours) in male stone-former, patient control and young adult control groups were 11.8, 10.2 and 10.8 respectively. The normal 24-hour magnesium excretions were similar to those given by Sutton and Watson (1969), while Evans et al. (1967) and Modlin (1967) obtained higher values for urinary magnesium
RELATIONSHIP OF URINARY CATIONS IN PATIENTS WITH RENAL CALCULI
24 1
in normal controls. According to the latter, there was no difference in the magnesium excretion between stone-formers and normal controls, but they found that stone-formers with normal calcium excretion showed a significant reduction in urinary magnesium. The results in this investigation suggest that no significant difference in magnesium excretion exists between any of the groups, in spite of the fact that most of the stone-formers are normocalciuric. With regard to the concentrations of calcium and magnesium, relative to creatinine, in early-morning urine, a moderately significant increase in calcium, and reduction in magnesium were observed for stoneformers compared with controls. These findings may be of some significance, as it is probable that the concentrated urine specimens are responsible for the deposition of stone-forming material. It has been reported by King et al. (1968) that the majority of stone-formers have an abnormally high calcium/magnesium ratio, calculated from 24-hour excretion values. When Oreopoulous el al. (1968) estimated the calcium/magnesium ratios from the calcium and magnesium concentrations in early-morning urine of stone-formers and controls, a significant increase in the mean calcium/magnesium ratio was observed in the stone-former's urine. The validity of ratios to relate 2 constituents and to compare these ratios between different groups have been criticised by Papadimitriou and Ram (1968) and Sutton and Watson (1968). In this study, the correlation coefficient and linear regression equation for 24-hour calcium on magnesium in each group were determined. The correlation was found to be highly significant in all groups, but there was no significant difference in the linear regression lines for calcium on magnesium between stoneformers and controls. With regard to the examination of early-morning urines, there was no significant difference in the calcium on magnesium linear regression equations for stone-formers and controls, in spite of the differences in calcium and magnesium concentrations. These results confirm the view that the use of ratios to compare excretory patterns between groups may be misleading. Modlin (1967) estimated sodium and calcium in white controls, Bantu and stone-formers. The Bantu, who are relatively free of stone disease, were reported to have a significantly higher excretion of sodium than normal white subjects, while stone-formers had a much lower excretion of sodium than normal controls. It was also stated that the mean sodium/calcium ratio was significantly lower in stone-formers than in white controls, and that the sodium/calcium ratio for the Bantu was much higher than the other two groups. On this evidence it was postulated that a high level of urinary sodium, and particularly an elevated sodium/calcium ratio, may prevent the occurrence of renal calculi. However, it has been found in this survey that stone-formers and normal controls excreted similar amounts of sodium per day. The young adult controls are exceptional in the fact that they excrete a greater amount of sodium, paralleling their increased excretion of other cations. The disparity between results for male patient controls (average age 47 years) and young healthy adults (average age 22 years) may be the reason for differences in interpretation of other workers. The relationships of calcium to sodium, expressed by the linear regression equations, were not significantly different. The urinary potassium excretion of male stone-formers was unexpectedly low in comparison with the values for both normal patients and young adults. There was no significant difference in the linear regression equations of stone-formers and patient controls, although it appears that the former have a lower excretion of potassium relative to calcium. There was no evidence to suggest that the potassium levels in the diet of stone-formers differs from that of the normal population. At present, no obvious explanation can be found for the low 24-hour potassium excretion of stone-formers. The early-morning urines of stone-formers and young adult controls have similar concentrations of sodium and potassium relative to creatinine.
Conclusion The 24-hour urinary excretions of calcium, magnesium and sodium by male idiopathic stoneformers do not appear to be significantly different from normal male controls. Potassium
242
BRITISH JOURNAL OF UROLOGY
excretion by stone-formers, however, was significantly lower than the controls. A moderately significant increase in calcium and a reduction in magnesium, relative to 1 g of creatinine, were observed in early-morning urines of patients with renal calculi. These abnormalities in concentrated urine samples which pass slowly through the renal tract overnight may be an important factor in the initiation and development of calcium-containing stones. Summary
Calcium, magnesium, sodium and potassium were estimated in the 24-hour urine collections of 101 idiopathic male stone-formers, 89 male patient controls and 59 young male adult controls. The results were calculated in terms of 24-hour volume and 1 g of creatinine. The concentrations of the 4 cations, relative to a gram of creatinine, were also determined in the early-morning urines of 41 male stone-formers and 43 young male adult controls. No difference was observed in the 24-hour excretion of calcium, magnesium and sodium between the stone-formers and controls. The mean daily potassium excretion, however, was significantly reduced in the urine of stone-formers. The linear regression equations were calculated for calcium on magnesium, calcium on sodium, and calcium on potassium, using the 24-hour excretion values of these cations. Only the calcium on potassium line of stone-formers was significantly different from that of the normal subjects. A significant increase by stone-formers in the urinary calcium concentration of their earlymorning specimens was found. The high concentration of urinary calcium in overnight urines of stone-formers combined with a low magnesium concentration might possibly contribute to the development of renal stone disease. The authors thank Dr J. D. Merrett, Department of Medical Statistics, Queen’s University, Belfast, for his help with the statistical analysis. They also wish to express their gratitude to Sister E. Cox and Sister I. Young for arranging the collection of specimens from their patients.
References BULUSU,L., HODGKINSON, A., NORDIN,B. E. C. and PEACOCK, M. (1970). Urinary excretion of calcium and creatinine in relation to age and body weight in normal subjects and patients with renal calculus. Clinical Science, 38, 601-612.
DAVIS,R. H., MORGAN, D. B. and RIVLIN,R. S. (1970). The excretion of calcium in the urine and its relation to calcium intake, sex and age. Clinical Science, 39, 1-12. EVANS,R. A., FORBES, M. A., SUTTON,R. A. L. and WATSON, L. (1967). Urinary excretion of calcium and magnesium in patients with calcium-containing renal stones.Lancet, 2,958-961. KINO,J. S. (1971). Currents in renal stone research. Clinical Chemistry, 17, 971-982. KING,J. S., O’CONNOR, F. J., SMITH,M. J. V. and CRouse, L. (1968). The urinary calcium/magnesium ratio in calcigerous stone formers. InvestigPtive Urology, 6, 60-65. MODLIN,M. (1967). The aetiology of renal stone: a new concept arising from studies on a stone-free population. Annals of the Royal College of Surgeons, 40,155-178. OREOPOULOS, D. G., SOYANNWO, M. A. and MCGEOWN,M. G. (1968). Magnesium/calcium ratio in urine of patients with renal stones. Lancet, 2,420-422. PAPADIMITRIOU, M. and RAM, M. D. (1968). Urinary magnesium and renal stones. Lancet, 2, 577. SUITON. R. A. L. and WATSON,L. (1969). Urinary excretion of calcium and magnesium in primary hyperparathyroidism. Lancet, i, 1000-1003. -(1968). Urinary magnesium and renal stones. Lancet, 2,636. WHITE,R. W., COHEN,R. D., VINCE,F. P., WILLIAMS, G., BLANDY, J. and TRESIDDER, G. C. (1969). Minerals in the urine of stone-formers and their spouses. In: Renal Stone Research Symposium, Ed. A. Hodgkinson and B. E. C. Nordin, London: Churchill, pp. 289-296.
The Authors S. G. Welshman, MSc, FRIC, Chief Clinical Biochemist. M. G. McGeown, MD, PhD,FRCPEd, Consultant Nephrologist.
