- Email: [email protected]
Normal values for random urinary calcium to creatinine ratios in infancy
Normal values for random urinary calcium to creatinine ratios in infancy James D. Sargent, MD, Therese A. Stukel, PhD, James Kresel, PhD, a n d Robert Z. Klein, MD From the Departments of Pediatrics, Community and FamityMedicine, and Pharmacy, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
Objective: To determine normal values for the urinary c a l c i u m / c r e a t i n i n e ratio (UCa/Cr) in infants. To assess the i m p a c t of short-term supplementation of infant formula with calcium and phosphorus on UCa/Cr in a g r o u p of infants. Design: We determined UCa/Cr in randomly c o l l e c t e d urine samples from a group of children and adults. Short-term supplementation of infant formula with calcium g l y c e r o p h o s p h a t e was carried out in 21 infants, and UCa/Cr was monitored in a before-and-after trial. Setting: A pediatric clinic at an a c a d e m i c center (infants and adults), and a day-care center (older children). Participants: A total of 103 infants between 5 days and 7 months of age, 40 infants between 8 and 17 months of age, 41 children between 18 months and 6 years, and 31 adults. Results: The 95th percentiles for molar UCa/Cr for the different age groups were as follows: less than 7 months, 2.42 (0~86 mg/mg); 7 to 18 months, 1.69 (0.60 m g / mg); 19 months to 6 years, 1.18 (0.42 m g / m g ) ; and adults, 0.61 (0.22 m g / m g ) . Regression analysis i n d i c a t e d a statistically significant decline in a v e r a g e UCa/Cr with age (R2 - 0.115, p <0.0004 for log [UCa/Cr] vs log [age]). The g e o m e t r i c means for the two groups of infants were significantly greater than those of the older children and the adults (p <0.05). Values for UCa/Cr in adults in our sample were c o m p a r a b l e to those previously reported. We d e t e c t e d no significant changes in mean UCa/Cr during week-long periods of c a l c i u m supplementation of up to 1.8 gm of c a l c i u m and 1.39 gm of phosphorus pe r liter of formula. Conclusion: We c o n c l u d e that normal values for UCa/Cr are much higher in infants than in older children and adults; UCa/Cr is age-related and declines g r a d u a l l y in the first several years of life, and short-term supplementation of infant formula with c a l c i u m g l y c e r o p h o s p h a t e has minimal effect on UCa/Cr. (J PEDIATR 1993;123:393-7)
Evidence that hypercalciuria with normal serum calcium concentrations is associated with hematuria 1, 2 and urolithSupported in part by the Hitchcock Foundation and by grant MCJ-330597 from the Maternal and Child Health Bureau (Title V, Social Security Act), Health Resources and Services Administration. Submitted for publication July 31, 1992; accepted May 7, t993. Reprint requests: James Sargent, MD, Pediatrics and Adolescent Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH 03756. Copyright | 1993 by Mosby-Year Book, Inc. 0022-3476/93/$l.00 + .10 9/20/48449
iasis in children 3 has prompted an interest in screening for hypercalciuria. Hypercalciuria in childhood has been defined as a urinary calcium excretion of more than 4 mg/kg body weight per 24 hours. 4, 5 Use of measured calcium exUCa/Cr
Ratio of urinary calcium to creatinine concentrations
creted per day has been superseded by measurement of the urinary calcium/creatinine ratio in random urine samples because of the difficulty in obtaining 24-hour urine samples; close correlation with 24-hour calcium excretion has been
393
394
Sargent et al.
The Journal of Pediatrics September 1993
T a b l e I. Descriptive statistics for random molar UCa/Cr* in children and adults A g e group
n
Geometric mean molar UCa/Cr
Median molar UCa/Cr
Interquartile range/2t
95th percentile
<7 mo (1) 7-18 mo (2) 19 mo-3 yr (3) Adult (4)
103 40 41 31
0.67 (0.24)$ 0.56 (0.20)5 0.30 (0.11) 0.27 (0.10)
0.83 (0.29) 0.58 (0.21) 0.35 (0.12) 0.32 (0.11)
0.51 (0.18) 0.42 (0.15) 0.26 (0.09) 0.12 (0.04)
2.42 (0.86) 1.69 (0.60) 1.18 (0.42) 0.61 (0.22)
*Reported as a molarratio (obtainedby dividingurinarycalciumin micromolesper literby urinarycreatininein mieromolesper liter), withthe milligram/milligram ratio followingin parentheses. tValue representshalf the differencebetweenthe 75th and 25th percentile,and is a nonparametricmeasureof the variabilityof a sample. ~:Thesemeanvaluesare significantlydifferentfrom those of groups 3 and 4 (p <0.05) by Tukeymultiplecomparison.
shown in adults, 6 healthy children,7, 8 and children with urolithiasis.9 We became interested in the issue of urinary calcium excretion when planning a trial of dietary calcium-phosphorus supplementation in children 3 to 18 months of age who were at risk for increased environmental lead exposure. We chose to use U C a / C r as a way of monitoring calcium excretion in the infants but were unable to find published data on U C a / C r in children of this age. The purpose of this study was to determine normal values for random U C a / C r in healthy infants and to compare these values with those of older children and adults. In addition, we determined the effects of short-term calcium-phosphorus supplementation on UCa/Cr. METHODS
Eligibility and study protocol. Children in the study population were drawn from the general pediatrics clinic at Dartmouth-Hitchcock Medical Center, which serves a predominantly white population in and around Lebanon, N.H. They were eligible for the study if their birth weight was >__2500gm, they had no major chronic illness, and they were being seen for well-child care or minor illness not affecting intake. Unless toilet trained, eligible children had adhesive urine collection bags attached by nurses preparing the child for the visit. If the child voided during the course of the visit, the urine was collected and frozen. Healthy older children were enrolled from the medical center day-care facility, and the adults were normal health care workers. Similar eligibility requirements applied to the 21 infants enrolling in the trial of calcium glycerophosphate supplementation, except that the age range was limited to 2 to 12 months. Moreover, the infants had to be consuming a commercial infant formula, and their parents had to be willing to change it to Enfamil Plus Iron (supplied by Mead Johnson & Co., Evansville, Ind.) for a 4-week period. In this trial, we supplemented the infant formula with calcium glycerophosphate in a stepwise fashion, as described below. I n the first week, infants consumed unsup-
plemented study formula containing 0.465 gm elemental calcium and 0.317 gm elemental phosphorus per liter. This regimen was followed by a 1-week regimen of consuming the base formula supplemented with calcium glycerophosphate to 1.2 gm elemental calcium and 0.92 gm elemental phosphorus per liter, and then a 2-week regimen of consuming formula supplemented to 1.8 gm elemental calcium and 1.39 gm elemental phosphorus per liter. Five urine specimens were collected, one at baseline and one after each week of study formula use. At each visit, parents were asked about tolerance to the feeding mixture, intake, and the development of side effects or acute illness. The parents supplemented the feeding mixture by adding a premeasured packet of calcium glycerophosphate (allowing for the hygroscopic property of glycerophosphate) to 26 ounces of formula prepared from concentrate. The supplemental calcium did not precipitate out of solution after 48 hours of refrigeration. The final solutions were found to have the desired concentrations of calcium by atomic absorption spectrophotometry. Characteristics of study population. Urine specimens were collected from 103 infants between 5 days and 7 months of age (group 1), 40 infants between 8 and 18 months (group 2), 41 children between 19 months and 6 years (group 3), and 31 adults (group 4). Complete information on age, milk or formula intake, and weight was available for 126 of 143 infants ~ 18 months of age, for whom mean weight and formula intake were 7.4 kg (range, 2.7 to 14 kg) and 898 ml/24 hr (range, 150 to 1680 ml/24 hr). Infant formula was supplemented with calcium glycerophosphate in a group of 21 infants whose ages ranged between 65 and 334 days. Urine was successfully obtained at each stage in 19 of the 21 infants. The fifth urine specimen from one of the patients was discarded at home and was uninterpretable for another because of stool contamination. Informed consent. Verbal informed consent was obtained to collect random urine samples, and written informed consent was obtained from parents in the calcium supplemen-
The Journal o f Pediatrics Volume 123, Number 3
Sargent et al.
395
Table II. Trial of calcium glycerophosphate supplementation: weight gain, and mean and maximum UCa/Cr*
Time
n
Calciumt (gm/L)
Phosphorus1" (gin/L)
Mean weight gain (kg)
Mean UCa/Cr
Maximum UCa/Cr
Baseline 21 ---0.82 (0.29) 1.6 (0.56) Week 1 21 0.465 0.317 0.11 0.75 (0.27) 1.8 (0.62) 2 21 1.2 0.92 0.09 0.78 (0.28) 2.1 (0.74) 3 21 1.8 1.39 0.11 0.86 (0.30) 2.3 (0.81) 4 19 1.8 1.39 0.15 0.98 (0.35) 2.3 (0.81) *Reported as a molarratio (obtainedby dividingurinarycalciumvalues,in micromolesper liter, by urinary creatininevalue, in micromolesper liter), with the milligram/milligramratio followingin parentheses. tin infant formula.
tation study. Studies and consent forms were approved by the human studies committee at the Dartmouth-Hitchcock Medical Center. Sample analysis. Urine specimens were analyzed in batches of 10. Urine creatinine concentration was determined on the Olympus Demand analyzer (Olympus Corp., Lake Success, N.Y.) by a modification of the Jaff6 reaction with the use of reagents from Worthington Diagnostics Systems, Inc. (Freehold, N.J.). Urinary calcium values were determined for the first 82 infants by using Sigma Diagnostics (St. Louis, Mo.) reagents, which utilize o-cresolphthalein complexone as a chromogenic reagent. Our laboratory subsequently changed the method of urinary calcium determination and began using the Ektachem technique (Eastman Kodak Co., Rochester, N.Y.), which employs Arsenazo III to produce a colored complex. Regression analysis for 20 urine calcium determinations by using both methods showed almost perfect correlation (R 2 = 0.993) and was used to correct the Sigma method to the Ektachem method. Seven duplicate samples analyzed in a blind fashion at the midpoint of the study showed good reproducibility; the average difference for duplicate samples was 2.4% of the mean value. All values for U C a / C r are reported as a molar ratio (obtained by dividing urinary calcium concentration in micromoles per liter by urinary creatinine concentration in micromoles per liter), with the milligram/milligram ratio following in parentheses. Statistical analysis. Age groups were based on the need for normal values in another study of calcium supplementation of infant formula, when urine collection began at 3 to 7 months of age and continued through 18 months of age. Multiple regression analysis was used to determine significant relationships among UCa/Cr, age, and weight. Because the UCa/Cr, age, and weight distributions were highly skewed, analyses were performed on the logarithmically transformed values of these variables. One-way analysis of variance, followed by the Tukey multiple comparison procedure, was used to assess differences in the geomet-
ric mean of U C a / C r values among the four age groups. Repeated-measures analysis was used to detect the change in mean U C a / C r according to the amount of supplemental calcium. The chi-square test was used to analyze changes in the rate of side effects reported during the trial. Analyses were performed by using generalized linear model (GLM) of the Statistical Analysis Systems statistical package (SAS Institute Inc., Cary, N.C.). RESULTS
Urinary calcium/creatinine ratio. Table I lists the median and geometric mean values of the UCa/Cr, the interquartile range (a nonparametric measure of variability useful for skewed data), and the 95th percentiles by age group. Both age and weight were inversely related to U C a / C r values. Age and weight were highly correlated, so we chose to present the relationship between age and U C a / C r to be consistent with other studies on the subject. The Figure shows a log-log plot of U C a / C r versus age; the regression line (log [UCa/Cr] = 0 . 1 8 - 0 . 1 9 log [age]) shows a statistically significant decline in average U C a / C r with age (R 2 = 0.115;p <0.0001). The geometric means of U C a / C r for the two infant groups (groups 1 and 2) were significantly higher (p <0.05) than the geometric means for older children and adults (groups 3 and 4). In addition, U C a / C r also became less variable with increasing age; the interquartile range became smaller (Table I). The estimate of the 95th percentile for UCa/Cr, or the upper limit of normal in these groups, was approximately four times higher for young infants (2.42) than for the adults (0.61). Trial of calcium glycerophosphate supplementation. Table II demonstrates that calcium supplementation had no effect on either mean weight gain or mean U C a / C r in these infants. Calcium supplementation was associated with a significantly increased number of reports of passage of hard stools compared with that during the week when patients received unsupplemented formula, especially in week 3, the first week of supplementation with 1.8 gm calcium per liter
396
Sargent et al.
The Journal of Pediatrics September 1993
10 O
g , t-
1 o
oo
0 O
0
'-'
.,-,o
0~[~
0.1
0
0
0
ao
0 0
0
0
~ 0
0 0 log(UCa/Cr) = 0.18 - 0.19 log(Age), R 2 = .115, p < .0001
0.01
9
i
I day
L
i
i iLlll
I
I0 days
i
t I
11
I00 days
3 years
I
I
!
30 years
Age Figure.
Log-log plot of UCa/Cr versus age. Age range, 5 days to 70 years; n = 215.
(p --- <0.05). Stool consistency appeared to return to normal the following week. No other side effects were reported. DISCUSSION The U C a / C r is higher in infants than in older children and adults, and steadily diminishes with age. We are unable to determine from our data when this ratio approaches that of adults. In our sample, the upper limit of normal for U C a / C r in the group aged ll/: to 6 years was twice that for adults, even though the geometric mean values were not statistically different; this may reflect a larger variance in the younger age group. The U C a / C r values for young infants and adults in this study are similar to those for representative groups from published reports. The median for young infants in this Sample is similar to d a t a published b y Rowe et al., 1~ in which median U C a / C r for growing premature infants between birth a n d 3 months was 2.3 (0.81). The mean U C a / C r for this adult sample (0.30 [0.11]) is similar to mean values reported by N0rdin 6 (0.35 [0.12]) and Pak et al. ll (0.16 [0.06] fasting; 0.37 [0.13] after l gm oral calcium load). Finally, the 95th percentile for U C a / C r in our adult sample (0.61 [0.22]) is very close to the commonly accepted upper limit of normal in the adult population (0.57
[o.20]). ~ Increased U C a / C r in young children may be due to less excretion of creatinine per unit of lean body mass in this age group. Creatinine excretion in adults <50 years of age is 20 to 25 mg/kg per day. Creatinine excretion in the first year
of life averages <10 mg/kg, with wide individual variation.12~4 Although the figures vary among reports, presumably because of the small numbers studied at different ages, excretion slowly increases to 20 mg/kg by adolescence, with an average excretion of about 14 mg/kg in children from 2 to 10 years of age. 1517 The literature is not entirely consistent with respect to changes in urinary creatinlne excretion with age, the one exception being the recent study by DeSanto et aI. ~8 showing greater creatinine excretion in younger children; these investigators did not study infants. Increased urinary calcium excretion also may have affected the U C a / C r in the infants in our study. Few data are available on urinary calcium excretion in infants except for the report by Rowe et al., 1~ cited above, in which the 95th percentile for 24-hour urine calcium excretion in healthy premature infants was 8.9 mg/kg. In a balance study, Fomon et al. 19 published data for 24-hour urine calCium excretion for 131 samples in 21 infants on various feeding regimens; their mean age was 89 days (range, 8 to 179 days). We calculated median 24-hour calcium excretion for these 131 samples and found it to be 5 mg/kg per day; the 95th percentile was 11 mg/kg per day, well above the upper limit 0f 4 mg/kg per day used for older children and for adults. Whatever the reasons for the inverse relationship between U C a / C r and age, it is not appropriate to use normal values for older children and adults in the screening or monitoring of infants. Clinicians should be aware of higher normal values for infants and young children when invest|-
The Journal o f Pediatrics Volume 123, Number 3
gating hematuria or nephrolithiasis, so that these children are not misclassified as having hypercalciuria and inappropriately given low-calcium diets or diuretic therapy. If possible, clinicians should confirm hypercalciuria with a timed urine specimen in a child whose random U C a / C r exceeds the normal limit. This study also demonstrates that short-term calcium glycerophosphate supplementation of infant formula does not lead to abnormally high urinary calcium excretion. W e did not determine the fate of the additional dietary calcium. Two balance studies of premature infants whose formula was supplemented with calcium and phosphate indicate increased calcium retention and increased stool excretion of calcium10, 2o; neither author found a relationship between urinary calcium excretion and calcium supplementation. Devizia et al. 21 found a significant increase in urinary calcium excretion by term infants whose formula was supplemented to 1024 mg calcium per liter (control formula, 389 m g / L ) . In contrast to our study, the De Vizia study maintained the phosphorus content of the calcium-supplemented formula at control levels. In our study the c a l c i u m / phosphorus ratio was similar for calcium-supplemented formula (1.3) and control formula (1.47). Maintaining a relatively low calcium/phosphorus ratio may thus have led to less urinary calcium excretion. Consistent with this hypothesis are findings by Knapp 5 that raising the dietary calcium/phosphorus ratio leads to an increase in urinary calcium excretion, and those by Widdowson et al., z2 who, in balance studies of normal infants, showed that phosphate supplementation of human milk reduced urinary excretion of calcium without affecting its absorption. W e conclude that normal values for random U C a / C r are much higher in infants and young children than in older children and adults. Our data indicate an inverse log-log relationship between U C a / C r and age. We suggest that clinicians use these normal values when screening infants for hypercalciuria or when monitoring infants with nephrolithiasis. Finally, normative data are lacking and should be obtained for timed urinary calcium excretion in children <3 years of age. We gratefully acknowledge the biostatistical support of Eugene Demidenko and Lee Mott. We thank Alyson Percy and the nursing staff in the pediatrics clinic at Dartmouth-Hitehcock Medical Center for their support in obtaining urine specimens, and Cecile Smith, Winnefred Breau, and Dr. Walter Noll, of the DartmouthHitchcock Medical Center Clinical Chemistry Laboratory, Department of Pathology. REFERENCES
1. Stapleton BF, Shane R, Noe HN, Jerkins G. Hypercalciuria in children with hematuria. N Engl J Med 1984;310:1345-8.
Sargent et al.
397
2. Stapleton FB, Southwest Pediatric Nephrology Study Group. Idiopathic hypercalciuria: association with isolated hematuria and risk for urolithiasis in children. Kidney Int 1990;37:80711. 3. Stapleton FB, Noe HN, Roy S. Hyperealeiuria in children with urolithiasis. Am J Dis Child 1982;136:675-8. 4. Royer P. Hereditary tubular diseases. In: Royer P, Habib R, Mathies H, Broyer M, eds. Pediatric nephrology. Philadelphia: WB Saunders, 1974:68. 5. Knapp EL. Factors influencing the urinary excretion of calcium. I. In normal persons. J Clin Invest 1947;45:182202. 6. Nordin BEC. Assessment of calcium excretion from the urinary calcium/ereatinine ratio. Lancet 1959;2:368-71. 7. Ring E, Borkenstein M. Use of the calcium-creatinine ratio in diagnosis and therapy. Pediatr Padol 1987;22:245-50. 8. Ghazali S, Barratt TM. Urinary excretion of calcium and magnesium in children. Arch Dis Child 1974;49:97-101. 9. Hymes LC, Warshaw BL. Idiopathic hypercalciuria: renal and absorptive subtypes in children. Am J Dis Child 1984;138:17680. 10. Rowe JC, Goetz CA, Carey DE, Horak E. Achievement of in utero retention of calcium and phosphorus accompanied by high calcium excretion in very low birth weight infants fed a fortified formula. J PEO~ATR1987;110:581-5. 11. Pak CY, Kaplan R, Bone H, Townsend J, Waters O. A simple test for the diagnosis of absorptive, resorptive and renal hypercalciurias. N Engl J Med 1975;292:497-500. 12. Catherwood R, Stearns G. Creatine and creatinine excretion in infancy. J Biochem 1937;119:201-14. 13. Daniels AL, Hejinian LM. Growth of infants from the standpoint of physical measurements and of nitrogen metabolism. Am J Dis Child 1929;38:499-506. 14. Marples E, Levine SZ. Creatinuria of infancy and childhood. Am J Dis Child 1936;51:30-57. 15. Aldert G, Bruschke G, Dietze F, Franke H, Haase J. Age-dependent changes and normal variations of creatine and creatinine excretion. Zeitsehrift fur Alternsforschung 1967;20: 113-8. 16. Jacobs DS, Kasten BL, Denott WR, Wolfson WL. Laboratory test handbook. St. Louis: Mosby, 1984. 17. Meites S. Pediatric clinical chemistry. Washington, D.C.: American Association of Clinical Chemistry, 1977:258. 18. De Santo NG, Di Iorio B, Capasso G, et al. Population-based data on urinary excretion of calcium, magnesium, oxalate, phosphate, and uric acid in children from Cimitile (southern Italy). Pediatr Nephrol 1992;6:149-57. 19. Fomon S J, Owen GM, Jensen RL, Thomas LN. Calcium and phosphorus balance studies with normal full-term infants fed pooled human milk or various formulas. Am J Clin Nutr 1963;12:346-56. 20. Giles MM, Fenton MH, Shaw B, et al. Sequential calcium and phosphorus balance studies in preterm infants. J PEDIATR 1987;110:591-8. 21. De Vizia B, Fomon S J, Nelson SE, Edwards BE, Ziegler EE. Effect of dietary calcium on metabolic balance of normal infants. Pediatr Res 1985;19:800-6. 22. Widdowson EM, McCance RA, Harrison GE, Sutton A. Effect of giving phosphate supplements to breast-fed babies on absorption and excretion of calcium, strontium, magnesium, and phosphorus. Lancet 1963;2:1250-1.
Objective: To determine normal values for the urinary c a l c i u m / c r e a t i n i n e ratio (UCa/Cr) in infants. To assess the i m p a c t of short-term supplementation of infant formula with calcium and phosphorus on UCa/Cr in a g r o u p of infants. Design: We determined UCa/Cr in randomly c o l l e c t e d urine samples from a group of children and adults. Short-term supplementation of infant formula with calcium g l y c e r o p h o s p h a t e was carried out in 21 infants, and UCa/Cr was monitored in a before-and-after trial. Setting: A pediatric clinic at an a c a d e m i c center (infants and adults), and a day-care center (older children). Participants: A total of 103 infants between 5 days and 7 months of age, 40 infants between 8 and 17 months of age, 41 children between 18 months and 6 years, and 31 adults. Results: The 95th percentiles for molar UCa/Cr for the different age groups were as follows: less than 7 months, 2.42 (0~86 mg/mg); 7 to 18 months, 1.69 (0.60 m g / mg); 19 months to 6 years, 1.18 (0.42 m g / m g ) ; and adults, 0.61 (0.22 m g / m g ) . Regression analysis i n d i c a t e d a statistically significant decline in a v e r a g e UCa/Cr with age (R2 - 0.115, p <0.0004 for log [UCa/Cr] vs log [age]). The g e o m e t r i c means for the two groups of infants were significantly greater than those of the older children and the adults (p <0.05). Values for UCa/Cr in adults in our sample were c o m p a r a b l e to those previously reported. We d e t e c t e d no significant changes in mean UCa/Cr during week-long periods of c a l c i u m supplementation of up to 1.8 gm of c a l c i u m and 1.39 gm of phosphorus pe r liter of formula. Conclusion: We c o n c l u d e that normal values for UCa/Cr are much higher in infants than in older children and adults; UCa/Cr is age-related and declines g r a d u a l l y in the first several years of life, and short-term supplementation of infant formula with c a l c i u m g l y c e r o p h o s p h a t e has minimal effect on UCa/Cr. (J PEDIATR 1993;123:393-7)
Evidence that hypercalciuria with normal serum calcium concentrations is associated with hematuria 1, 2 and urolithSupported in part by the Hitchcock Foundation and by grant MCJ-330597 from the Maternal and Child Health Bureau (Title V, Social Security Act), Health Resources and Services Administration. Submitted for publication July 31, 1992; accepted May 7, t993. Reprint requests: James Sargent, MD, Pediatrics and Adolescent Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH 03756. Copyright | 1993 by Mosby-Year Book, Inc. 0022-3476/93/$l.00 + .10 9/20/48449
iasis in children 3 has prompted an interest in screening for hypercalciuria. Hypercalciuria in childhood has been defined as a urinary calcium excretion of more than 4 mg/kg body weight per 24 hours. 4, 5 Use of measured calcium exUCa/Cr
Ratio of urinary calcium to creatinine concentrations
creted per day has been superseded by measurement of the urinary calcium/creatinine ratio in random urine samples because of the difficulty in obtaining 24-hour urine samples; close correlation with 24-hour calcium excretion has been
393
394
Sargent et al.
The Journal of Pediatrics September 1993
T a b l e I. Descriptive statistics for random molar UCa/Cr* in children and adults A g e group
n
Geometric mean molar UCa/Cr
Median molar UCa/Cr
Interquartile range/2t
95th percentile
<7 mo (1) 7-18 mo (2) 19 mo-3 yr (3) Adult (4)
103 40 41 31
0.67 (0.24)$ 0.56 (0.20)5 0.30 (0.11) 0.27 (0.10)
0.83 (0.29) 0.58 (0.21) 0.35 (0.12) 0.32 (0.11)
0.51 (0.18) 0.42 (0.15) 0.26 (0.09) 0.12 (0.04)
2.42 (0.86) 1.69 (0.60) 1.18 (0.42) 0.61 (0.22)
*Reported as a molarratio (obtainedby dividingurinarycalciumin micromolesper literby urinarycreatininein mieromolesper liter), withthe milligram/milligram ratio followingin parentheses. tValue representshalf the differencebetweenthe 75th and 25th percentile,and is a nonparametricmeasureof the variabilityof a sample. ~:Thesemeanvaluesare significantlydifferentfrom those of groups 3 and 4 (p <0.05) by Tukeymultiplecomparison.
shown in adults, 6 healthy children,7, 8 and children with urolithiasis.9 We became interested in the issue of urinary calcium excretion when planning a trial of dietary calcium-phosphorus supplementation in children 3 to 18 months of age who were at risk for increased environmental lead exposure. We chose to use U C a / C r as a way of monitoring calcium excretion in the infants but were unable to find published data on U C a / C r in children of this age. The purpose of this study was to determine normal values for random U C a / C r in healthy infants and to compare these values with those of older children and adults. In addition, we determined the effects of short-term calcium-phosphorus supplementation on UCa/Cr. METHODS
Eligibility and study protocol. Children in the study population were drawn from the general pediatrics clinic at Dartmouth-Hitchcock Medical Center, which serves a predominantly white population in and around Lebanon, N.H. They were eligible for the study if their birth weight was >__2500gm, they had no major chronic illness, and they were being seen for well-child care or minor illness not affecting intake. Unless toilet trained, eligible children had adhesive urine collection bags attached by nurses preparing the child for the visit. If the child voided during the course of the visit, the urine was collected and frozen. Healthy older children were enrolled from the medical center day-care facility, and the adults were normal health care workers. Similar eligibility requirements applied to the 21 infants enrolling in the trial of calcium glycerophosphate supplementation, except that the age range was limited to 2 to 12 months. Moreover, the infants had to be consuming a commercial infant formula, and their parents had to be willing to change it to Enfamil Plus Iron (supplied by Mead Johnson & Co., Evansville, Ind.) for a 4-week period. In this trial, we supplemented the infant formula with calcium glycerophosphate in a stepwise fashion, as described below. I n the first week, infants consumed unsup-
plemented study formula containing 0.465 gm elemental calcium and 0.317 gm elemental phosphorus per liter. This regimen was followed by a 1-week regimen of consuming the base formula supplemented with calcium glycerophosphate to 1.2 gm elemental calcium and 0.92 gm elemental phosphorus per liter, and then a 2-week regimen of consuming formula supplemented to 1.8 gm elemental calcium and 1.39 gm elemental phosphorus per liter. Five urine specimens were collected, one at baseline and one after each week of study formula use. At each visit, parents were asked about tolerance to the feeding mixture, intake, and the development of side effects or acute illness. The parents supplemented the feeding mixture by adding a premeasured packet of calcium glycerophosphate (allowing for the hygroscopic property of glycerophosphate) to 26 ounces of formula prepared from concentrate. The supplemental calcium did not precipitate out of solution after 48 hours of refrigeration. The final solutions were found to have the desired concentrations of calcium by atomic absorption spectrophotometry. Characteristics of study population. Urine specimens were collected from 103 infants between 5 days and 7 months of age (group 1), 40 infants between 8 and 18 months (group 2), 41 children between 19 months and 6 years (group 3), and 31 adults (group 4). Complete information on age, milk or formula intake, and weight was available for 126 of 143 infants ~ 18 months of age, for whom mean weight and formula intake were 7.4 kg (range, 2.7 to 14 kg) and 898 ml/24 hr (range, 150 to 1680 ml/24 hr). Infant formula was supplemented with calcium glycerophosphate in a group of 21 infants whose ages ranged between 65 and 334 days. Urine was successfully obtained at each stage in 19 of the 21 infants. The fifth urine specimen from one of the patients was discarded at home and was uninterpretable for another because of stool contamination. Informed consent. Verbal informed consent was obtained to collect random urine samples, and written informed consent was obtained from parents in the calcium supplemen-
The Journal o f Pediatrics Volume 123, Number 3
Sargent et al.
395
Table II. Trial of calcium glycerophosphate supplementation: weight gain, and mean and maximum UCa/Cr*
Time
n
Calciumt (gm/L)
Phosphorus1" (gin/L)
Mean weight gain (kg)
Mean UCa/Cr
Maximum UCa/Cr
Baseline 21 ---0.82 (0.29) 1.6 (0.56) Week 1 21 0.465 0.317 0.11 0.75 (0.27) 1.8 (0.62) 2 21 1.2 0.92 0.09 0.78 (0.28) 2.1 (0.74) 3 21 1.8 1.39 0.11 0.86 (0.30) 2.3 (0.81) 4 19 1.8 1.39 0.15 0.98 (0.35) 2.3 (0.81) *Reported as a molarratio (obtainedby dividingurinarycalciumvalues,in micromolesper liter, by urinary creatininevalue, in micromolesper liter), with the milligram/milligramratio followingin parentheses. tin infant formula.
tation study. Studies and consent forms were approved by the human studies committee at the Dartmouth-Hitchcock Medical Center. Sample analysis. Urine specimens were analyzed in batches of 10. Urine creatinine concentration was determined on the Olympus Demand analyzer (Olympus Corp., Lake Success, N.Y.) by a modification of the Jaff6 reaction with the use of reagents from Worthington Diagnostics Systems, Inc. (Freehold, N.J.). Urinary calcium values were determined for the first 82 infants by using Sigma Diagnostics (St. Louis, Mo.) reagents, which utilize o-cresolphthalein complexone as a chromogenic reagent. Our laboratory subsequently changed the method of urinary calcium determination and began using the Ektachem technique (Eastman Kodak Co., Rochester, N.Y.), which employs Arsenazo III to produce a colored complex. Regression analysis for 20 urine calcium determinations by using both methods showed almost perfect correlation (R 2 = 0.993) and was used to correct the Sigma method to the Ektachem method. Seven duplicate samples analyzed in a blind fashion at the midpoint of the study showed good reproducibility; the average difference for duplicate samples was 2.4% of the mean value. All values for U C a / C r are reported as a molar ratio (obtained by dividing urinary calcium concentration in micromoles per liter by urinary creatinine concentration in micromoles per liter), with the milligram/milligram ratio following in parentheses. Statistical analysis. Age groups were based on the need for normal values in another study of calcium supplementation of infant formula, when urine collection began at 3 to 7 months of age and continued through 18 months of age. Multiple regression analysis was used to determine significant relationships among UCa/Cr, age, and weight. Because the UCa/Cr, age, and weight distributions were highly skewed, analyses were performed on the logarithmically transformed values of these variables. One-way analysis of variance, followed by the Tukey multiple comparison procedure, was used to assess differences in the geomet-
ric mean of U C a / C r values among the four age groups. Repeated-measures analysis was used to detect the change in mean U C a / C r according to the amount of supplemental calcium. The chi-square test was used to analyze changes in the rate of side effects reported during the trial. Analyses were performed by using generalized linear model (GLM) of the Statistical Analysis Systems statistical package (SAS Institute Inc., Cary, N.C.). RESULTS
Urinary calcium/creatinine ratio. Table I lists the median and geometric mean values of the UCa/Cr, the interquartile range (a nonparametric measure of variability useful for skewed data), and the 95th percentiles by age group. Both age and weight were inversely related to U C a / C r values. Age and weight were highly correlated, so we chose to present the relationship between age and U C a / C r to be consistent with other studies on the subject. The Figure shows a log-log plot of U C a / C r versus age; the regression line (log [UCa/Cr] = 0 . 1 8 - 0 . 1 9 log [age]) shows a statistically significant decline in average U C a / C r with age (R 2 = 0.115;p <0.0001). The geometric means of U C a / C r for the two infant groups (groups 1 and 2) were significantly higher (p <0.05) than the geometric means for older children and adults (groups 3 and 4). In addition, U C a / C r also became less variable with increasing age; the interquartile range became smaller (Table I). The estimate of the 95th percentile for UCa/Cr, or the upper limit of normal in these groups, was approximately four times higher for young infants (2.42) than for the adults (0.61). Trial of calcium glycerophosphate supplementation. Table II demonstrates that calcium supplementation had no effect on either mean weight gain or mean U C a / C r in these infants. Calcium supplementation was associated with a significantly increased number of reports of passage of hard stools compared with that during the week when patients received unsupplemented formula, especially in week 3, the first week of supplementation with 1.8 gm calcium per liter
396
Sargent et al.
The Journal of Pediatrics September 1993
10 O
g , t-
1 o
oo
0 O
0
'-'
.,-,o
0~[~
0.1
0
0
0
ao
0 0
0
0
~ 0
0 0 log(UCa/Cr) = 0.18 - 0.19 log(Age), R 2 = .115, p < .0001
0.01
9
i
I day
L
i
i iLlll
I
I0 days
i
t I
11
I00 days
3 years
I
I
!
30 years
Age Figure.
Log-log plot of UCa/Cr versus age. Age range, 5 days to 70 years; n = 215.
(p --- <0.05). Stool consistency appeared to return to normal the following week. No other side effects were reported. DISCUSSION The U C a / C r is higher in infants than in older children and adults, and steadily diminishes with age. We are unable to determine from our data when this ratio approaches that of adults. In our sample, the upper limit of normal for U C a / C r in the group aged ll/: to 6 years was twice that for adults, even though the geometric mean values were not statistically different; this may reflect a larger variance in the younger age group. The U C a / C r values for young infants and adults in this study are similar to those for representative groups from published reports. The median for young infants in this Sample is similar to d a t a published b y Rowe et al., 1~ in which median U C a / C r for growing premature infants between birth a n d 3 months was 2.3 (0.81). The mean U C a / C r for this adult sample (0.30 [0.11]) is similar to mean values reported by N0rdin 6 (0.35 [0.12]) and Pak et al. ll (0.16 [0.06] fasting; 0.37 [0.13] after l gm oral calcium load). Finally, the 95th percentile for U C a / C r in our adult sample (0.61 [0.22]) is very close to the commonly accepted upper limit of normal in the adult population (0.57
[o.20]). ~ Increased U C a / C r in young children may be due to less excretion of creatinine per unit of lean body mass in this age group. Creatinine excretion in adults <50 years of age is 20 to 25 mg/kg per day. Creatinine excretion in the first year
of life averages <10 mg/kg, with wide individual variation.12~4 Although the figures vary among reports, presumably because of the small numbers studied at different ages, excretion slowly increases to 20 mg/kg by adolescence, with an average excretion of about 14 mg/kg in children from 2 to 10 years of age. 1517 The literature is not entirely consistent with respect to changes in urinary creatinlne excretion with age, the one exception being the recent study by DeSanto et aI. ~8 showing greater creatinine excretion in younger children; these investigators did not study infants. Increased urinary calcium excretion also may have affected the U C a / C r in the infants in our study. Few data are available on urinary calcium excretion in infants except for the report by Rowe et al., 1~ cited above, in which the 95th percentile for 24-hour urine calcium excretion in healthy premature infants was 8.9 mg/kg. In a balance study, Fomon et al. 19 published data for 24-hour urine calCium excretion for 131 samples in 21 infants on various feeding regimens; their mean age was 89 days (range, 8 to 179 days). We calculated median 24-hour calcium excretion for these 131 samples and found it to be 5 mg/kg per day; the 95th percentile was 11 mg/kg per day, well above the upper limit 0f 4 mg/kg per day used for older children and for adults. Whatever the reasons for the inverse relationship between U C a / C r and age, it is not appropriate to use normal values for older children and adults in the screening or monitoring of infants. Clinicians should be aware of higher normal values for infants and young children when invest|-
The Journal o f Pediatrics Volume 123, Number 3
gating hematuria or nephrolithiasis, so that these children are not misclassified as having hypercalciuria and inappropriately given low-calcium diets or diuretic therapy. If possible, clinicians should confirm hypercalciuria with a timed urine specimen in a child whose random U C a / C r exceeds the normal limit. This study also demonstrates that short-term calcium glycerophosphate supplementation of infant formula does not lead to abnormally high urinary calcium excretion. W e did not determine the fate of the additional dietary calcium. Two balance studies of premature infants whose formula was supplemented with calcium and phosphate indicate increased calcium retention and increased stool excretion of calcium10, 2o; neither author found a relationship between urinary calcium excretion and calcium supplementation. Devizia et al. 21 found a significant increase in urinary calcium excretion by term infants whose formula was supplemented to 1024 mg calcium per liter (control formula, 389 m g / L ) . In contrast to our study, the De Vizia study maintained the phosphorus content of the calcium-supplemented formula at control levels. In our study the c a l c i u m / phosphorus ratio was similar for calcium-supplemented formula (1.3) and control formula (1.47). Maintaining a relatively low calcium/phosphorus ratio may thus have led to less urinary calcium excretion. Consistent with this hypothesis are findings by Knapp 5 that raising the dietary calcium/phosphorus ratio leads to an increase in urinary calcium excretion, and those by Widdowson et al., z2 who, in balance studies of normal infants, showed that phosphate supplementation of human milk reduced urinary excretion of calcium without affecting its absorption. W e conclude that normal values for random U C a / C r are much higher in infants and young children than in older children and adults. Our data indicate an inverse log-log relationship between U C a / C r and age. We suggest that clinicians use these normal values when screening infants for hypercalciuria or when monitoring infants with nephrolithiasis. Finally, normative data are lacking and should be obtained for timed urinary calcium excretion in children <3 years of age. We gratefully acknowledge the biostatistical support of Eugene Demidenko and Lee Mott. We thank Alyson Percy and the nursing staff in the pediatrics clinic at Dartmouth-Hitehcock Medical Center for their support in obtaining urine specimens, and Cecile Smith, Winnefred Breau, and Dr. Walter Noll, of the DartmouthHitchcock Medical Center Clinical Chemistry Laboratory, Department of Pathology. REFERENCES
1. Stapleton BF, Shane R, Noe HN, Jerkins G. Hypercalciuria in children with hematuria. N Engl J Med 1984;310:1345-8.
Sargent et al.
397
2. Stapleton FB, Southwest Pediatric Nephrology Study Group. Idiopathic hypercalciuria: association with isolated hematuria and risk for urolithiasis in children. Kidney Int 1990;37:80711. 3. Stapleton FB, Noe HN, Roy S. Hyperealeiuria in children with urolithiasis. Am J Dis Child 1982;136:675-8. 4. Royer P. Hereditary tubular diseases. In: Royer P, Habib R, Mathies H, Broyer M, eds. Pediatric nephrology. Philadelphia: WB Saunders, 1974:68. 5. Knapp EL. Factors influencing the urinary excretion of calcium. I. In normal persons. J Clin Invest 1947;45:182202. 6. Nordin BEC. Assessment of calcium excretion from the urinary calcium/ereatinine ratio. Lancet 1959;2:368-71. 7. Ring E, Borkenstein M. Use of the calcium-creatinine ratio in diagnosis and therapy. Pediatr Padol 1987;22:245-50. 8. Ghazali S, Barratt TM. Urinary excretion of calcium and magnesium in children. Arch Dis Child 1974;49:97-101. 9. Hymes LC, Warshaw BL. Idiopathic hypercalciuria: renal and absorptive subtypes in children. Am J Dis Child 1984;138:17680. 10. Rowe JC, Goetz CA, Carey DE, Horak E. Achievement of in utero retention of calcium and phosphorus accompanied by high calcium excretion in very low birth weight infants fed a fortified formula. J PEO~ATR1987;110:581-5. 11. Pak CY, Kaplan R, Bone H, Townsend J, Waters O. A simple test for the diagnosis of absorptive, resorptive and renal hypercalciurias. N Engl J Med 1975;292:497-500. 12. Catherwood R, Stearns G. Creatine and creatinine excretion in infancy. J Biochem 1937;119:201-14. 13. Daniels AL, Hejinian LM. Growth of infants from the standpoint of physical measurements and of nitrogen metabolism. Am J Dis Child 1929;38:499-506. 14. Marples E, Levine SZ. Creatinuria of infancy and childhood. Am J Dis Child 1936;51:30-57. 15. Aldert G, Bruschke G, Dietze F, Franke H, Haase J. Age-dependent changes and normal variations of creatine and creatinine excretion. Zeitsehrift fur Alternsforschung 1967;20: 113-8. 16. Jacobs DS, Kasten BL, Denott WR, Wolfson WL. Laboratory test handbook. St. Louis: Mosby, 1984. 17. Meites S. Pediatric clinical chemistry. Washington, D.C.: American Association of Clinical Chemistry, 1977:258. 18. De Santo NG, Di Iorio B, Capasso G, et al. Population-based data on urinary excretion of calcium, magnesium, oxalate, phosphate, and uric acid in children from Cimitile (southern Italy). Pediatr Nephrol 1992;6:149-57. 19. Fomon S J, Owen GM, Jensen RL, Thomas LN. Calcium and phosphorus balance studies with normal full-term infants fed pooled human milk or various formulas. Am J Clin Nutr 1963;12:346-56. 20. Giles MM, Fenton MH, Shaw B, et al. Sequential calcium and phosphorus balance studies in preterm infants. J PEDIATR 1987;110:591-8. 21. De Vizia B, Fomon S J, Nelson SE, Edwards BE, Ziegler EE. Effect of dietary calcium on metabolic balance of normal infants. Pediatr Res 1985;19:800-6. 22. Widdowson EM, McCance RA, Harrison GE, Sutton A. Effect of giving phosphate supplements to breast-fed babies on absorption and excretion of calcium, strontium, magnesium, and phosphorus. Lancet 1963;2:1250-1.