BODY SIZE plays an important role in nephrolithiasis.

Transcription

1 Body Size and 24-Hour Urine Composition Eric N. Taylor, MD, and Gary C. Curhan, MD, ScD Background: Greater body mass index (BMI) is a risk factor for kidney stones. However, the relation between BMI and the urinary excretion of many lithogenic factors remains unclear. Methods: We studied urine ph, urine volume, and 24-hour urinary excretion of calcium, oxalate, citrate, uric acid, sodium, magnesium, potassium, phosphate, and creatinine in stone-forming and non stone-forming participants in the Health Professionals Follow-Up Study (599 stone-forming and 404 non stone-forming men), Nurses Health Study (888 stone-forming and 398 non stoneforming older women), and Nurses Health Study II (689 stone-forming and 295 non stone-forming younger women). Each cohort was divided into quintiles of BMI. Tests of linear trend were conducted by 1-way analysis of variance. Linear regression models were adjusted for age, history of stone disease, dietary intake, and urinary factors. Results: Participants with greater BMIs excreted more urinary oxalate (P for trend < 0.04), uric acid (P < 0.001), sodium (P < 0.001), and phosphate (P < 0.001) than participants with lower BMIs. There was an inverse relation between BMI and urine ph (P < 0.02). Positive associations between BMI and urinary calcium excretion in men and stone-forming younger women (P < 0.02) did not persist after adjustment for urinary sodium and phosphate excretion. Because of differences in urinary volume and excretion of inhibitors such as citrate, we observed no relation between BMI and urinary supersaturation of calcium oxalate. Urinary supersaturation of uric acid increased with BMI (P < 0.01). Conclusion: Positive associations between BMI and urinary calcium excretion likely are due to differences in animal protein and sodium intake. The greater incidence of kidney stones in the obese may be due to an increase in uric acid nephrolithiasis. Am J Kidney Dis 48: by the National Kidney Foundation, Inc. INDEX WORDS: Body mass index; body size; urinary composition; urinary calcium; urinary ph; kidney stones; nephrolithiasis. BODY SIZE plays an important role in nephrolithiasis. Greater body mass index (BMI), greater weight, larger waist circumference, and weight gain are associated independently with increased risk for kidney stone formation. 1 However, the mechanisms underlying the relation between larger body size and increased stone risk are unknown. Prior data suggest that urinary excretion of many stone promoters or inhibitors may vary with body size. Examination of more than 4,500 patients with a history of kidney stones showed that urinary ph, an important risk factor for uric acid nephrolithiasis, was related inversely to body weight. 2 In another study of nearly 6,000 stone formers, individuals weighing more than 120 kg excreted more urinary calcium, oxalate, and uric acid (all factors associated with calcium oxalate stone risk) than individuals weighing less than 100 kg. 3 However, because of differences in urine volume, heavier and lighter participants had similar concentrations of urinary calcium and oxalate. A study of more than 500 calcium oxalate stone formers showed a positive association between BMI and urinary oxalate level in women and between BMI and urinary calcium level in men. 4 However, this study included only 52 obese stone formers, of whom only 17 were women. Another study of stone formers reported a greater prevalence of hypercalciuria, hyperoxaluria, and hyperuricosuria in 83 obese individuals compared with their nonobese counterparts. 5 Because prior studies included only stoneforming subjects, 2-5 included limited categories of body size, 3,5 studied relatively few participants with obesity, 4,5 or did not measure differences in dietary intake, 3-5 the effect of body size on the excretion of many urinary factors remains unclear. To evaluate the relation between body size and urine composition, we studied urinary ph, urine volume, and 24-hour urinary excretion From the Renal Division and Channing Laboratory, Department of Medicine, Brigham and Women s Hospital, Harvard Medical School; and Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, MA. Received July 10, 2006; accepted in revised form September 11, Originally published online as doi: /j.ajkd on November 6, Support: Grants DK73381, DK59583, CA87969, CA55075, and CA50385 from the National Institutes of Health. Potential conflicts of interest: None. Address reprint requests to Eric N. Taylor, MD, Channing Laboratory, Third Floor, Brigham and Women s Hospital, 181 Longwood Ave, Boston, MA entaylor@ partners.org 2006 by the National Kidney Foundation, Inc /06/ $32.00/0 doi: /j.ajkd American Journal of Kidney Diseases, Vol 48, No 6 (December), 2006: pp

2 906 of calcium, oxalate, citrate, uric acid, sodium, magnesium, potassium, phosphate, and creatinine in a subset of participants with and without kidney stone disease in the Health Professionals Follow-Up Study (HPFS), the Nurses Health Study I (NHS I), and the Nurses Health Study II (NHS II). METHODS Study Population Health Professionals Follow-Up Study. In 1986, a total of 51,529 male dentists, optometrists, osteopaths, pharmacists, podiatrists, and veterinarians between the ages of 40 and 75 years completed and returned an initial questionnaire that provided detailed information on medical history, lifestyle, and medications. This cohort, like NHS I and NHS II, is followed up by means of biennial mailed questionnaires that include inquiries about the incidence of newly diagnosed diseases, including kidney stones. Nurses Health Study I. In 1976, a total of 121,700 female registered nurses between the ages of 30 and 55 years enrolled in NHS I by completing and returning an initial questionnaire. Nurses Health Study II. In 1989, a total of 116,671 female registered nurses between the ages of 25 and 42 years enrolled in NHS II by completing and returning an initial questionnaire. The current analysis was restricted to whites because few minorities provided urine samples. Ascertainment of Body Size For each cohort, information for weight and height was obtained on the baseline questionnaire. Self-reported weight was updated every 2 years. BMI was calculated as weight in kilograms divided by the square of height in meters. Selfreported weight was validated in HPFS and NHS I. 6 Selfreported weights from 123 men and 140 women in the 2 cohorts correlated highly with values obtained by technicians who visited participants at home (r 0.97 for men and women). 6 Ascertainment of Other Covariates Information for age was obtained on the baseline questionnaire. Thiazide diuretic use was updated every 2 years. Information for hypertension and diabetes mellitus was obtained at baseline and then every 2 years. The validity of these self-reported diseases was documented. 7-9 A semiquantitative food-frequency questionnaire asked about the average use of more than 130 foods and beverages during the previous year. In addition, respondents provided information on the use of nutritional supplements, either alone or in multivitamin form. A version of this foodfrequency questionnaire is mailed to study participants every 4 years. The reproducibility and validity of the foodfrequency questionnaires in the HPFS and NHS I were 10, 11 documented. Nutrient intake was computed from the reported frequency of consumption of each specified unit of food and TAYLOR AND CURHAN from US Department of Agriculture data for the content of the relevant nutrient in specified portions. Intake of supplemental vitamins and minerals in multivitamin or isolated form was determined by brand, type, and frequency of reported use. Participants reported on the interval diagnosis of kidney stones every 2 years. Any study participant who reported a new kidney stone was sent an additional questionnaire to determine the date of occurrence and symptoms from the stone. Medical record review confirmed 96% of cases in NHS I, 98% in NHS II, and 95% in HPFS. Urine Collection Procedure Twenty-four hour urine samples from participants with a history of kidney stones and randomly selected controls were collected in 2 cycles as part of an ongoing study to compare the urine composition of those with and without kidney stone disease. In the first cycle, which spanned from 1994 to 1999, we obtained one 24-hour urine collection from stone formers and randomly selected non stone formers (396 NHS I participants, 199 NHS II participants, and 451 HPFS participants provided urine). In the second cycle, which started in 2001, we asked additional stone formers and randomly selected non stone formers for two 24-hour urine collections. In the first cycle, participants were ineligible if they were older than 70 years in HPFS or older than 65 years in NHS I or had a history of cancer or cardiovascular disease. In the second cycle, participants were ineligible if they were older than 75 years or had a history of cancer (other than nonmelanoma skin cancer). Response rates were reported previously 12 and were similar in stone formers and non stone formers. There were no substantial differences between participants who collected and did not collect urine in age, BMI, and intake of calcium, sodium, and animal protein. In this study, 1,328 NHS I participants, 989 NHS II participants, and 1,019 HPFS participants provided at least one 24-hour urine collection. Twenty-four hour urine collections were performed using the system provided by Mission Pharmacal (San Antonio, TX). We sent a kit containing all necessary supplies, including a 4-L jug with a marker-impregnated sponge attached to the bottom, to participants agreeing to collect a 24-hour specimen. The jugs also contained preservative to prevent bacterial growth. To assist in urine collection, female participants were sent specially designed hats and male participants were sent urinals. Upon completion of the collection, participants poured samples into 2 small vials, 1 of which contained acid preservative. The vials were returned to Mission Pharmacal in a prepaid self-addressed Federal Express mailer. The concentration of the marker was used to calculate the total volume of the collection. Analytic Procedures Used for Urine Measurements Calcium and magnesium were measured by using an atomic absorption spectrophotometer. Creatinine, uric acid, citrate, and phosphorus were measured by means of a Cobas centrifugal analyzer (Roche, Montclair, NJ). Oxalate was analyzed by means of ion chromatography. Sodium and potassium were determined directly by using flame emission

3 BODY SIZE AND URINE COMPOSITION 907 photometry. Urine ph was measured by using a ph electrode. We previously sent blinded split samples to assess reproducibility: coefficients of variation for all factors analyzed were less than 10%. Statistical Analysis In the primary analysis, we included participants who provided a single 24-hour urine collection. If a participant submitted more than 1 urine collection, we used the first sample. To exclude those with incomplete collections, we limited the analysis to participants with 24-hour urine creatinine values greater than 600 mg for women and 800 mg for men (42 NHS I participants, 5 NHS II participants, and 16 HPFS participants were excluded by using this criterion). Participants with uric acid crystals noted in the 24-hour urine collection (54 participants in NHS I, 51 participants in NHS II, and 88 participants in HPFS) were included, except in analyses of urinary uric acid and urinary supersaturation of uric acid. In secondary analyses, we restricted the study to participants who submitted two 24-hour urine collections. In these analyses, participants were included only if the creatinine value in each urine collection differed by less than 30% from the mean creatinine value of the 2 collections. Values for each urinary factor were obtained by calculating the arithmetic mean of both collections. Participants in each cohort were separated into 5 categories according to quintiles of BMI. Analyses also were performed by using quintiles of weight. Data from stone formers and non stone formers in each cohort were analyzed separately. Tests of linear trend were conducted by 1-way analysis of variance. To examine the independent relation between each urinary factor and categories of body size, we used linear regression to adjust simultaneously for multiple covariates. Variables considered in multivariate analyses were age, kidney stone history, intake of such dietary factors as animal protein and calcium, urinary ph, urine volume, and 24-hour urinary excretion of calcium, oxalate, citrate, uric acid, sodium, magnesium, potassium, and phosphate. All P are 2 tailed. We calculated 95% confidence intervals (CIs) for estimates obtained in linear regression analyses. All data were analyzed using SAS software, version 9.1 (SAS Institute Inc, Cary, NC). The research protocol for this study was reviewed and approved by the Institutional Review Board of Brigham and Women s Hospital (Boston, MA). RESULTS Characteristics of men, older women, and younger women according to quintile of BMI are listed in Table 1. Frequencies of hypertension, diabetes, and thiazide use increased with increasing BMI. Other differences in demographic characteristics across quintiles of BMI were not consistent from group to group. In all 3 populations, energy-adjusted animal protein intake increased with increasing BMI. Because of a lower frequency of supplemental calcium use, older women with a greater BMI consumed less calcium than those with a lower BMI. Tables 2, 3, and 4 list mean values for each urinary factor by quintile of BMI for men, older women, and younger women. Multivariate linear regression results comparing differences in urinary factors by quintile of BMI are listed in Table 5. Table 1. Demographic and Dietary Factors by Quintile of BMI in Men (HPFS), Older Women (NHS I), and Younger Women (NHS II) Quintile of BMI HPFS Q1/Q2/Q3/Q4/Q5 NHS I Q1/Q2/Q3/Q4/Q5 NHS II Q1/Q2/Q3/Q4/Q5 No. of patients 200/194/215/195/ /263/250/259/ /196/195/199/197 Median BMI (kg/m 2 ) 22.3/24.3/25.8/27.4/ /23.8/26.5/29.2/ /23.2/25.6/29.5/37.1 Age (y) 63.9/63.7/63.6/64.0/ /66.5/65.8/66.7/ /48.2/49.2/48.3/48.5 Hypertension (%) 22/27/32/42/56 32/40/43/57/70 5/14/16/25/39 Diabetes (%) 3/5/5/7/10 3/4/9/13/19 2/1/2/4/11 Thiazide use (%) 5/9/7/12/13 7/14/12/13/23 3/5/7/7/12 Menopausal (%) Not applicable 97/97/95/99/96 22/28/33/31/35 Supplement use (%) 56/53/47/47/49 50/57/48/57/50 56/47/57/49/50 Multivitamins Calcium 13/13/12/15/11 58/55/50/46/43 36/33/41/27/33 Vitamin D 2/1/2/2/1 29/33/26/34/33 6/3/5/3/2 Nutrient intake* Animal protein (g/d) 56/59/57/61/62 46/48/50/50/51 50/52/55/56/57 Calcium (mg/d) 908/939/900/926/918 1,216/1,243/1,115/1,105/1,102 1,149/1,170/1,196/1,041/1,129 NOTE. Values expressed as means unless otherwise noted. To convert calcium in mg to mmol, multiply by Abbreviation: Q, quintile. *Energy adjusted.

4 908 TAYLOR AND CURHAN Table 2. Twenty-Four Hour Urine Composition by Quintile of BMI in Stone-Forming and Non Stone-Forming Men, HPFS Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 No. of Patients (NSF/SF) 76/124 80/114 84/131 90/105 74/125 P for Trend Creatinine (mg) NSF 1,427 1,550 1,598 1,700 1, SF 1,489 1,634 1,675 1,763 1, Calcium (mg) NSF SF Oxalate (mg) NSF SF Citrate (mg) NSF SF Uric acid (mg) NSF SF Sodium (meq) NSF SF Potassium (meq) NSF SF Magnesium (mg) NSF SF Phosphate (mg) NSF ,045 1,075 1, SF 970 1,086 1,078 1,148 1, ph (units) NSF SF Volume (L) NSF SF Relative supersaturation Calcium oxalate NSF SF Brushite NSF SF Uric acid NSF SF NOTE. To convert creatinine in mg to mol, multiply by 8.84; calcium in mg to mmol, multiply by ; oxalate in mg to mol, multiply by 11.1; citrate in mg to mol, multiply by 5.205; uric acid in mg to mol, multiply by 5.948; sodium and potassium in meq to mmol, multiply by 1.0; magnesium in mg to mmol, multiply by ; phosphate in mg to mmol, multiply by Abbreviations: SF, stone forming; NSF, non stone forming. Calcium In men and stone-forming younger women, but not older women, urinary calcium excretion increased with increasing BMI (P for trend 0.02). The positive association between BMI and urinary calcium excretion in men and younger women did not persist after adjustment for urinary phosphate and sodium excretion. In multivariate

5 BODY SIZE AND URINE COMPOSITION 909 Table 3. Twenty-Four Hour Urine Composition by Quintile of BMI in Stone-Forming and Non Stone-Forming Older Women, NHS I Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 No. of Patients (NSF/SF) 111/146 90/173 75/175 68/191 54/203 P for Trend Creatinine (mg) NSF ,036 1,148 1, SF ,065 1,099 1, Calcium (mg) NSF SF Oxalate (mg) NSF SF Citrate (mg) NSF SF Uric acid (mg) NSF SF Sodium (meq) NSF SF Potassium (meq) NSF SF Magnesium (mg) NSF SF Phosphate (mg) NSF SF ph (units) NSF SF Volume (L) NSF SF Relative supersaturation Calcium oxalate NSF SF Brushite NSF SF Uric acid NSF SF NOTE. To convert creatinine in mg to mol, multiply by 8.84; calcium in mg to mmol, multiply by ; oxalate in mg to mol, multiply by 11.1; citrate in mg to mol, multiply by 5.205; uric acid in mg to mol, multiply by 5.948; sodium and potassium in meq to mmol, multiply by 1.0; magnesium in mg to mmol, multiply by ; phosphate in mg to mmol, multiply by Abbreviations: SF, stone forming; NSF, non stone forming. analysis, an inverse relation between BMI and urinary calcium excretion was seen in older women. Older women in the highest quintile of BMI excreted 24.1 mg (0.60 mmol; 95% CI, 7.7 to 40.6 mg [0.19 to 1.01 mmol]) less urinary calcium per day than women in the lowest quintile of BMI.

6 910 TAYLOR AND CURHAN Table 4. Twenty-Four Hour Urine Composition by Quintile of BMI in Stone-Forming and Non Stone-Forming Younger Women, NHS II Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 No. of Patients (NSF/SF) 62/135 60/136 74/121 55/144 44/153 P for Trend Creatinine (mg) NSF 1,086 1,139 1,203 1,252 1, SF 1,045 1,179 1,224 1,274 1, Calcium (mg) NSF SF Oxalate (mg) NSF SF Citrate (mg) NSF SF Uric acid (mg) NSF SF Sodium (meq) NSF SF Potassium (meq) NSF SF Magnesium (mg) NSF SF Phosphate (mg) NSF SF , ph (units) NSF SF Volume (L) NSF SF Relative supersaturation Calcium oxalate NSF SF Brushite NSF SF Uric acid NSF SF NOTE. To convert creatinine in mg to mol, multiply by 8.84; calcium in mg to mmol, multiply by ; oxalate in mg to mol, multiply by 11.1; citrate in mg to mol, multiply by 5.205; uric acid in mg to mol, multiply by 5.948; sodium and potassium in meq to mmol, multiply by 1.0; magnesium in mg to mmol, multiply by ; phosphate in mg to mmol, multiply by Abbreviations: SF, stone forming; NSF, non stone forming. Oxalate BMI was associated positively with urinary oxalate excretion in all 3 cohorts (P for trend 0.04). After adjusting for urinary phosphate and uric acid excretion, this relation persisted only in older and younger women. Older and younger women in the highest quintile of BMI excreted 3.6 mg (40.0 mol; 95% CI, 1.9 to 5.3 mg [21.1

7 BODY SIZE AND URINE COMPOSITION 911 Table 5. Multivariate Linear Regression Results for 24-Hour Urine Composition, Presented as Differences in Urinary Excretion Between Lowest and Higher Quintiles of BMI HPFS NHS I NHS II Calcium (mg) BMI Quintile ( ) 11.2 ( ) 0.7 ( ) Q3 6.6 ( ) 5.2 ( ) 13.5 ( ) Q4 0.8 ( ) 20.8 ( 36.8 to 4.7) 13.2 ( ) Q5 2.7 ( ) 24.1 ( 40.6 to 7.7) 10.4 ( ) Oxalate (mg) BMI Quintile ( ) 0.3 ( ) 0.7 ( ) Q3 1.1 ( ) 0.4 ( ) 2.5 ( ) Q4 2.0 ( ) 1.0 ( ) 1.5 ( ) Q5 1.0 ( ) 3.6 ( ) 5.5 ( ) Citrate (mg) BMI Quintile 2 21 ( 74-31) 1 ( 49-46) 29 ( 28-86) Q3 15 ( 67-36) 12 ( 36-61) 6 ( 64-53) Q4 4 ( 58-50) 36 ( 14-85) 22 ( 36-81) Q5 64 ( 118 to 9) 24 ( 27-75) 43 ( ) Uric acid (mg) BMI Quintile ( ) 10.9 ( ) 0.6 ( ) Q ( ) 22.7 ( ) 20.2 ( ) Q ( ) 22.1 ( ) 54.9 ( ) Q ( ) 34.0 ( ) 53.8 ( ) Sodium (meq) BMI Quintile ( ) 2.6 ( ) 0.1 ( ) Q ( ) 1.3 ( ) 9.9 ( ) Q ( ) 13.3 ( ) 14.5 ( ) Q ( ) 28.1 ( ) 40.1 ( ) Potassium (meq) BMI Quintile ( ) 0.2 ( ) 1.6 ( ) Q3 2.1 ( ) 0.6 ( ) 0.8 ( ) Q4 4.2 ( 8.0 to 0.3) 2.2 ( ) 1.5 ( ) Q5 2.8 ( ) 0.5 ( ) 1.2 ( ) Magnesium (mg) BMI Quintile ( ) 0.1 ( ) 1.4 ( ) Q3 4.1 ( ) 2.5 ( ) 6.2 ( ) Q4 5.7 ( ) 5.3 ( ) 1.4 ( ) Q5 0.8 ( ) 6.2 ( ) 1.1 ( ) Phosphate (mg) BMI Quintile ( ) 17.2 ( ) 27.0 ( ) Q ( ) 27.2 ( ) 45.7 ( ) Q ( ) 66.8 ( ) 30.1 ( ) Q ( ) 46.8 ( ) 66.4 ( ) ph (units) BMI Quintile ( ) 0.05 ( ) 0.01 ( ) Q ( 0.18 to 0.01) 0.12 ( 0.20 to 0.04) 0.10 ( 0.18 to 0.01) Q ( ) 0.15 ( 0.23 to 0.07) 0.16 ( 0.24 to 0.07) Q ( 0.26 to 0.09) 0.17 ( 0.26 to 0.09) 0.23 ( 0.32 to 0.14) Volume (L) BMI Quintile ( ) 0.07 ( ) 0.08 ( ) Q ( ) 0.09 ( ) 0.14 ( ) Q ( ) 0.06 ( ) 0.08 ( ) Q ( ) 0.00 ( ) 0.08 ( ) NOTE. Lowest quintile is referent. Adjusted for case status, age, and quartile of each urinary factor except creatinine. Statistically significant values are in bold font. 95% confidence intervals are in parentheses. To convert creatinine in mg to mol, multiply by 8.84; calcium in mg to mmol, multiply by ; oxalate in mg to mol, multiply by 11.1; citrate in mg to mol, multiply by 5.205; uric acid in mg to mol, multiply by 5.948; sodium and potassium in meq to mmol, multiply by 1.0; magnesium in mg to mmol, multiply by ; phosphate in mg to mmol, multiply by Abbreviation: Q, BMI quintile.

8 912 to 58.8 mol]) and 5.5 mg (61.1 mol; 95% CI, 3.6 to 7.4 mg [40.0 to 82.1 mol]) more urinary oxalate per day than those in the lowest quintile of BMI, respectively. Citrate Urinary citrate excretion increased with BMI in stone-forming older and younger women (P for trend 0.05). In multivariate analyses, urinary citrate excretion was associated with BMI only in men. Men in the highest quintile of BMI excreted 64 mg (333 mol; 95% CI, 9 to 118 mg [47 to 614 mol]) less citrate per day than those in the lowest quintile of BMI. Uric Acid Urinary excretion of uric acid increased with increasing BMI (P for trend 0.001). In multivariate analyses, men, older women, and younger women in the highest quintile of BMI excreted 62.9 mg (374 mol; 95% CI, 26.6 to 99.1 mg [158 to 589 mol]), 34.0 mg (202 mol; 95% CI, 9.6 to 58.4 mg [57 to 347 mol]), and 53.8 mg (320 mol; 95% CI, 24.6 to 83.1 mg [146 to 494 mol]) more urinary uric acid per day than those in the lowest quintile of BMI, respectively. Sodium Urinary excretion of sodium increased with increasing BMI (P for trend 0.001). In multivariate analyses, men, older women, and younger women in the highest quintile of BMI excreted 31.5 meq (31.5 mmol; 95% CI, 20.0 to 43.0 meq), 28.1 meq (28.1 mmol; 95% CI, 19.0 to 37.1 meq), and 40.1 meq (40.1 mmol; 95% CI, 28.8 to 51.5 meq) more urinary sodium per day than those in the lowest quintile of BMI, respectively. Potassium Urinary potassium excretion increased with quintile of BMI only in stone-forming older and younger women (P for trend 0.05). In multivariate analyses, urinary potassium excretion was not associated with BMI. TAYLOR AND CURHAN Magnesium Urinary magnesium excretion increased with increasing BMI only in stone-forming men and younger women (P for trend 0.03). In multivariate analyses, urinary magnesium excretion was not associated with BMI. Phosphate Urinary phosphate excretion increased with BMI (P for trend 0.001). In multivariate analyses, men, older women, and younger women in the highest quintile of BMI excreted mg (4.0 mmol; 95% CI, 77.3 to mg [2.5 to 5.6 mmol]), 46.8 mg (1.5 mmol; 95% CI, 12.1 to 81.5 mg [0.4 to 2.6 mmol]), and 66.4 mg (2.1 mmol; 95% CI, 19.6 to mg [0.6 to 3.7 mmol]) more urinary phosphate per day than those in the lowest quintile of BMI, respectively. ph Urinary ph decreased with BMI (P for trend 0.02). In multivariate analyses, urinary ph values in men, older women, and younger women in the highest quintile of BMI were 0.18 (95% CI, 0.09 to 0.26), 0.17 (95% CI, 0.09 to 0.26), and 0.23 (95% CI, 0.14 to 0.32) less than those in the lowest quintile of BMI, respectively. Volume Although urine volume of participants in the highest quintile of BMI tended to be greater than that for participants in the lowest quintile, test for trend was statistically significant only in non stone-forming men (P 0.005). In multivariate analyses, urine volume was not associated with BMI. Supersaturation of Calcium Oxalate Relative urinary supersaturation of calcium oxalate did not increase across quintiles of BMI in any cohort. Supersaturation of Brushite Relative urinary supersaturation of brushite (calcium phosphate) decreased in older and younger women with increasing BMI (P for trend 0.03). There was no relation between BMI and relative urinary supersaturation of brushite in men. Supersaturation of Uric Acid Relative urinary supersaturation of uric acid increased with increasing BMI (P for trend 0.04).

9 BODY SIZE AND URINE COMPOSITION 913 Restriction of analyses to those with 2 urine collections, exclusion of participants administered thiazide diuretics, and exclusion of participants with a history of diabetes did not materially change results. Additional adjustment for smoking, hypertension, use of hormone replacement therapy (NHS I and NHS II), menopausal status (NHS II), oral contraceptive use (NHS II), and calcium and animal protein intake also did not change results. Multivariate analyses of the relation between quintiles of body weight (rather than BMI) and urinary composition yielded similar results, with the exception of urinary oxalate excretion in men. After adjusting for age, kidney stone history, and other urinary factors, men, older women, and younger women in the highest weight quintile (respective median weights, 100, 93, and 102 kg) excreted 2.6 mg (28.9 mol; 95% CI, 0.2 to 5.0 mg [2.2 to 55.5 mol]), 4.8 mg (53.3 mol; 95% CI, 3.1 to 6.6 mg [34.4 to 73.3 mol]), and 5.4 mg (59.9 mol; 95% CI, 3.4 to 7.3 mg [37.7 to 81.0 mol]) more urinary oxalate per day than those in the lowest weight quintile (respective median weights, 69, 56, and 55 kg). DISCUSSION We observed numerous and marked differences in the urine composition of both stoneforming and non stone-forming individuals with larger compared to smaller body size. Participants with greater BMIs excreted more urinary oxalate, uric acid, sodium, and phosphate than participants with lower BMIs and had lower urinary ph values. We observed a positive association between BMI and urinary calcium excretion only in men and stone-forming younger women. These unadjusted urine data are of primary interest when evaluating nephrolithiasis or other diseases in which both urine composition and body size may affect risk (such as osteoporosis and hypertension). Actual concentrations of urinary stone promoters and inhibitors affect kidney stone formation. 13 However, the adjusted analyses in our study may provide insight into the contribution of dietary factors to urine composition. For example, the positive relation between BMI and urinary calcium excretion in men and stoneforming younger women did not persist after adjustment for urinary phosphate and sodium excretion. Prior population-based studies noted a positive association between urinary calcium excretion and body size, 3-5 although 1 study found this relation to be true only in men. 4 Because obesity is associated with insulin resistance and insulin may increase renal fractional excretion of calcium, it would be reasonable to speculate that insulin resistance accounts for the relation between body size and urinary calcium excretion. However, in our study, participants with lower BMIs consumed less animal protein and less sodium than participants with higher BMIs, and prior data showed that restriction of dietary animal protein and sodium markedly decreased urinary calcium excretion. 17 In all 3 cohorts, beef and other foods high in animal protein represented a major contributor to total dietary and, presumably, urinary phosphate. Thus, the null association between body size and urinary calcium excretion in our adjusted analyses suggests that the positive association observed in prior studies may have been caused by dietary differences rather than fundamental changes in renal physiology mediated by greater BMI. Because large epidemiologic studies showed a marked increase in stone risk with increasing BMI, 1 because the majority of stone formers form calcium oxalate stones, 13 and because greater BMI is associated with increased urinary excretion of calcium, oxalate, and uric acid, we expected to find an increase in relative urinary supersaturation of calcium oxalate across quintiles of increasing BMI. However, because of differences in urinary volume and urinary excretion of such inhibitors as citrate, we observed no relation between BMI and urinary supersaturation of calcium oxalate in any of our study populations. To the extent that urinary supersaturation predicts risk for nephrolithiasis, our data suggest that risk for calcium oxalate stone formation does not increase with increasing body size. However, formulas used to calculate relative supersaturation of calcium salts do not account for some urinary inhibitors, such as phytate, that theoretically could vary among individuals of different body size. We observed a greater relative urinary supersaturation of uric acid with increasing BMI, largely because of the inverse relation between body size and urinary ph. The inverse relation

10 914 between body size and urinary ph was shown previously 2 and may be secondary to insulin resistance, which decreases renal ammonia excretion and impairs hydrogen ion buffering. In vitro and in vivo studies showed that insulin plays an important role in the renal production of ammonia, 18,19 and a metabolic trial in humans showed that insulin has a powerful effect on renal ammonium excretion. 20 Our results suggest that the increase in stone risk in the overweight and obese may represent an increase in incidence of uric acid nephrolithiasis. Of note, prior studies suggested that the prevalence of uric acid stones is greater in obese compared to nonobese subjects. 5,21 Some associations between body size and urine composition were observed only in stone formers. For example, the increases in urinary citrate and potassium excretion observed with increasing quintiles of BMI in stone-forming older and younger women were not seen in older and younger women without a history of kidney stones. It is possible that this may reflect differences in treatment in larger compared to smaller stone formers. However, information collected from incident cases in these cohorts showed that the majority of first-time stone formers did not undergo metabolic evaluation. 12 In addition, prior results from 24-hour urine collections showed expected differences between cases and controls, suggesting that dietary and/or pharmacological intervention was unlikely in the majority of firsttime stone-forming subjects. 12 Thus, differences observed between stone formers and non stone formers may reflect fundamental differences in physiological characteristics, rather than treatment. Our results also may have implications for patients with osteoporosis and hypertension. Greater bone mass and lower frequency of osteoporosis are associated with greater BMI In our study, urinary calcium excretion increased in men and younger women with increasing BMI despite similar calcium intake across BMI categories, suggesting greater net calcium losses in the overweight and obese. This counterintuitive finding may support the theory that greater levels of animal protein intake, as observed in individuals in our study with greater BMIs, promote intestinal absorption of dietary calcium, 26 or this may indicate an important role for dietary animal TAYLOR AND CURHAN protein mediated increases in factors that promote bone formation, such as insulin-like growth factor Hypertension also is associated positively with BMI, and it has been proposed that increased urinary calcium excretion is a central feature of essential hypertension. 28,29 However, the positive relation between urinary calcium excretion and BMI in our study suggests that previous population-based studies showing a relation between hypertension and urinary calcium excretion may have been confounded by body size. Of note, recent data from the cohorts in our study showed that urinary calcium levels were not related consistently to prevalent hypertension. 30 Limitations of our study deserve mention. Weight and height were self-reported. However, self-reported weight and height were validated in these cohorts. Other sources of error include overcollection and undercollection of 24-hour urine specimens. However, there is little reason to believe that participants with larger body size, as a group, would submit overcollections or undercollections of 24-hour urine samples. Furthermore, the observed relation between BMI and urinary creatinine excretion was as expected. Finally, the generalizability of our results may be limited. We do not have urine collections from younger men, and our study did not include urine collections from nonwhite participants. In conclusion, our data show multiple and marked differences in the urine composition of individuals with larger compared to smaller body size. Research examining the effect of body size on urine calcium excretion must account for differences in sodium and animal protein intake between larger and smaller individuals. In addition, because the relations between such physiological parameters as body size and the excretion of some urinary factors vary by stone-forming status, population-based studies of urine composition need to include adequate numbers of non stone-forming participants. Furthermore, the relation between excretion of many urinary factors and body size varies by age and sex. Finally, urinary supersaturation of calcium oxalate did not increase with BMI despite increases in urinary calcium, oxalate, and uric acid excretion. Although it is possible that formulas used to calculate urinary supersaturation of calcium salts do not include important factors that vary by

11 BODY SIZE AND URINE COMPOSITION 915 body size, our data suggest that the greater incidence of kidney stones in overweight and obese individuals may be due to an increase in uric acid, rather than calcium, nephrolithiasis. ACKNOWLEDGMENT The authors thank the study participants and Meir J. Stampfer, MD, DrPH, Walter C. Willett, MD, DrPH, Elaine M. Coughlan, Christine Jones, and Adam Summerfield. REFERENCES 1. Taylor EN, Stampfer MJ, Curhan GC: Obesity, weight gain, and the risk of kidney stones. JAMA 293: , Maalouf NM, Sakhaee K, Parks JH, Coe FL, Adams- Huet B, Pak CY: Association of urinary ph with body weight in nephrolithiasis. Kidney Int 65: , Powell CR, Stoller ML, Schwartz BF, et al: Impact of body weight on urinary electrolytes in urinary stone formers. Urology 55: , Siener R, Glatz S, Nicolay C, Hesse A: The role of overweight and obesity in calcium oxalate stone formation. Obes Res 12: , Ekeruo WO, Tan YH, Young MD, et al: Metabolic risk factors and the impact of medical therapy on the management of nephrolithiasis in obese patients. J Urol 172: , Rimm EB, Stampfer MJ, Colditz GA, Chute CG, Litin LB, Willett WC: Validity of self-reported waist and hip circumferences in men and women. Epidemiology 1: , Colditz G, Marin P, Stampfer M, et al: Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol 123: , Manson JE, Rimm EB, Stampfer MJ, et al: Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet 338: , Hu FB, Leitzmann MF, Stampfer MJ, Colditz GA, Willett WC, Rimm EB: Physical activity and television watching in relation to risk for type 2 diabetes mellitus in men. Arch Intern Med 161: , Willett WC, Sampson L, Stampfer MJ, et al: Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 122:51-65, Rimm EB, Giovannucci EL, Stampfer MJ, Colditz GA, Litin LB, Willett WC: Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol 135: , Curhan GC, Willett WC, Speizer FE, Stampfer MJ: Twenty-four-hour urine chemistries and the risk of kidney stones among women and men. Kidney Int 59: , Coe FL, Parks JH, Asplin JR: The pathogenesis and treatment of kidney stones. N Engl J Med 327: , Kerstetter J, Caballero B, O Brien K, Wurtman R, Allen L: Mineral homeostasis in obesity: Effects of euglycemic hyperinsulinemia. Metabolism 40: , Shimamoto K, Higashiura K, Nakagawa M, et al: Effects of hyperinsulinemia under the euglycemic condition on calcium and phosphate metabolism in non-obese normotensive subjects. Tohoku J Exp Med 177: , Nowicki M, Kokot F, Surdacki A: The influence of hyperinsulinaemia on calcium-phosphate metabolism in renal failure. Nephrol Dial Transplant 13: , Borghi L, Schianchi T, Meschi T, et al: Comparison of two diets for the prevention of recurrent stones in idiopathic hypercalciuria. N Engl J Med 346:77-84, Krivosikova Z, Spustova V, Dzurik R: Participation of P-dependent and P-independent glutaminases in rat kidney ammoniagenesis and their modulation by metabolic acidosis, hippurate and insulin. Physiol Res 47: , Chobanian MC, Hammerman MR: Insulin stimulates ammoniagenesis in canine renal proximal tubular segments. Am J Physiol 253:F1171-F1177, Abate N, Chandalia M, Cabo-Chan AV Jr, Moe OW, Sakhaee K: The metabolic syndrome and uric acid nephrolithiasis: Novel features of renal manifestation of insulin resistance. Kidney Int 65: , Daudon M, Lacour B, Jungers P: Influence of body size on urinary stone composition in men and women. Urol Res 34: , Reid IR, Ames R, Evans MC, et al: Determinants of total body and regional bone mineral density in normal postmenopausal women A key role for fat mass. J Clin Endocrinol Metab 75:45-51, Compston JE, Bhambhani M, Laskey MA, Murphy S, Khaw KT: Body composition and bone mass in postmenopausal women. Clin Endocrinol (Oxf) 37: , Khosla S, Atkinson EJ, Riggs BL, Melton LJ III: Relationship between body composition and bone mass in women. J Bone Miner Res 11: , Barrera G, Bunout D, Gattas V, de la Maza MP, Leiva L, Hirsch S: A high body mass index protects against femoral neck osteoporosis in healthy elderly subjects. Nutrition 20: , Kerstetter JE, O Brien KO, Insogna KL: Dietary protein, calcium metabolism, and skeletal homeostasis revisited. Am J Clin Nutr 78:S584-S592, 2003 (suppl 3) 27. Dawson-Hughes B: Interaction of dietary calcium and protein in bone health in humans. J Nutr 133:S852- S854, 2003 (suppl 3) 28. Cappuccio FP, Kalaitzidis R, Duneclift S, Eastwood JB: Unravelling the links between calcium excretion, salt intake, hypertension, kidney stones and bone metabolism. J Nephrol 13: , Oshima T, Young EW: Systemic and cellular calcium metabolism and hypertension. Semin Nephrol 15: , Taylor EN, Mount DB, Forman JP, Curhan GC: Association of prevalent hypertension with 24-hour urinary excretion of calcium, citrate, and other factors. Am J Kidney Dis 47: , 2006