CITRATE IS ATRICARBOXYLIC acid normally

Causes of Hypocitraturia in Recurrent Calcium Stone Formers: Focusing on Urinary Potassium Excretion Somnuek Domrongkitchaiporn, MD, Wasana Stitchantrakul, MSc, and Wachira Kochakarn, MD Background: Multiple factors associated with hypocitraturia have been identified. However, limited studies addressing the causal relationship to hypocitraturia are available. We therefore conducted this study to determine factors associated with hypocitraturia and show their causal relationship in recurrent calcium stone formers. Methods: Dietary review and 24-hour urine samples were obtained from all recurrent calcium stone formers referred for metabolic workup in the stone clinic. One month of oral potassium chloride supplementation was prescribed to stone formers to determine the causal relationship between urinary potassium and citrate levels. Results: Eighty-three subjects, 44 men and 39 women, were recruited to participate in this study. Hypocitraturia (citrate < 3 mg/d [<1.43 mmol/d]) was found in 5.6% of subjects. Four independent urinary variables associated with hypocitraturia were identified, including potassium level, net gastrointestinal alkaline absorption, calcium level, and titratable acid. Urinary potassium level was the strongest predictor of urinary citrate level. Hypocitraturic subjects also had lower fruit intake compared with subjects with high urinary citrate levels. Potassium chloride supplementation to a subgroup of this population (n 58) resulted in a significant increase in urinary citrate excretion (35.73 27.25 versus 34.15 3. mg/d [1.67.13 versus 1.45.14 mmol/d]; P <.2), but no alteration in fractional excretion of citrate (19.7% 2.7% versus 23.1% 2.4%; P >.5). Conclusion: Hypocitraturia was found to be a common risk factor associated with recurrent calcium stone formation and low urinary potassium level, low alkaline absorption, low urinary calcium level, and high titratable acid excretion. Hypocitraturia is predominantly of dietary origin. Estimation of fruit intake should be included in the metabolic evaluation for recurrent calcium stone formation. Am J Kidney Dis 48:546-554. 26 by the National Kidney Foundation, Inc. INDEX WORDS: Hypocitraturia; nephrolithiasis; potassium; alkaline absorption; calcium; titratable acid. From the Departments of Medicine and Surgery and Research Center, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand. Received December 8, 25; accepted in revised form June 12, 26. Originally published online as doi:1.153/j.ajkd.26.6.8 on August 15, 26. Support: None. Potential conflicts of interest: None. Address reprint requests to Somnuek Domrongkitchaiporn, MD, Department of Medicine, Ramathibodi Hospital, Mahidol University, Rama 6, Bangkok 14, Thailand. E-mail: rasdr@mahidol.ac.th 26 by the National Kidney Foundation, Inc. 272-6386/6/484-3$32./ doi:1.153/j.ajkd.26.6.8 CITRATE IS ATRICARBOXYLIC acid normally excreted in urine. Urinary citrate acts as an inhibitor for calcium oxalate stone formation through several mechanisms. First, urinary citrate binds to urinary calcium, forming a soluble complex and decreasing available free ionic calcium for calcium oxalate stone formation. 1 Urinary citrate also interacts with calcium oxalate crystal as an inhibitor of crystal aggregation 2,3 and crystal growth. 4 Low urinary citrate excretion is a well-accepted risk factor for calcium stone formation. 5 Increased urinary citrate excretion in hypocitraturic calcium stone formers by means of alkaline therapy was recommended to decrease the risk for recurrent calcium stone formation. 6 The incidence of hypocitraturia among calcium stone formers from various studies varies from 2% to 6%. 6 Several causes of hypocitraturia were identified, including renal tubular acidosis, 7 chronic diarrhea and malabsorption, 8 metabolic acidosis, 9 potassium deficiency, 1 low intestinal alkaline absorption, 11 low urinary calcium level, 12 and low urine volume. 13 In addition, a diet rich in animal protein 14 or low in vegetable fiber 13 may be associated with hypocitraturia. Frequencies of these causes of hypocitraturia in calcium stone formers may vary among different populations. However, most previous studies tended to focus on each individual cause of hypocitraturia, rather than perform multifactorial analysis on the relative impact of these factors. 12 A very high incidence of hypocitraturia was reported in the northeastern part of Thailand, ranging from 4% to 7%. 15,16 Hypocitraturia probably is the most common risk factor found in recurrent calcium stone formers. However, causes of hypocitraturia were never addressed systematically. Given a high incidence of hypocitraturia in our calcium stone formers, we therefore conducted this study to determine the relative influence of multiple associated risk factors for hy- 546 American Journal of Kidney Diseases, Vol 48, No 4 (October), 26: pp 546-554

CAUSES OF HYPOCITRATURIA 547 pocitraturia and show their causal relationships to hypocitraturia. METHODS Initial Evaluation Subjects were enrolled from recurrent calcium stone formers attending the nephrolithiasis clinic for metabolic evaluation of underlying causes of recurrent stone formation between July 22 and July 23. All subjects were residents of the central part of Thailand and referred from either the Department of Surgery, Division of Urology, Ramathibodi Hospital, or nearby hospitals. Pediatric patients ( 14 years) were excluded from this study. The diagnosis of recurrent stone formation was based on a history of at least 2 episodes of stone formation diagnosed by either spontaneous passing of stone or demonstration of new opaque stone on abdominal radiograph. Compositional analysis of renal stones was based on infrared spectroscopy of spontaneous passing stones, extracorporeal shock-wave lithotripsy, or surgical stone elimination. For episodes in which stones were not available for analysis, the diagnosis of calcium stone was based on abdominal radiograph and ultrasound examination. Patients with the following conditions were excluded from this study: staghorn calculi, failure to eliminate urinary tract infection before the study, impaired renal function (serum creatinine 1.5 mg/dl [ 132.6 mol/l]), polycystic kidney disease, malformation of the urological system, chronic diarrhea, urinary tract obstruction, renal tubular acidosis, and other systemic diseases that might affect calcium and bone metabolism. Renal tubular acidosis was excluded if a patient could acidify his urine (urinary ph 5.35) within 8 hours after.1 g/kg of ammonium chloride loading. 7 Patients administered medications that might have an effect on electrolyte, calcium, and phosphate levels and acid-base balance also were excluded from this study. If a subject underwent surgery or extracorporeal shock-wave lithotripsy for stone removal, the study was postponed for at least 2 months. One hundred two calcium stone formers attended our nephrolithiasis clinic. Nineteen patients were excluded, including 7 patients with renal tubular acidosis, 3 patients with polycystic kidney disease, 1 patient with medullary sponge kidney, 4 patients with staghorn calculi, 1 patient with single kidney, and 3 patients who failed to follow the study protocol. Only 83 patients, 44 men and 39 women, were eligible for this study. Mean age was 49.5 1.4 years, and mean body weight was 62.7 1. kg. All subjects, following a free-choice diet, were instructed by trained staff to perform two 24-hour urinary collections. Urine samples were kept under mineral oil. Toluene also was added to the collecting vessel as a preservative. The entire urine sample was kept in a refrigerator during collection and transferred to our laboratory the same day the urine collection was completed. Bacterial cultures were performed for all urine samples. Urine samples with heavy bacterial contamination ( 1 5 colonies/ml), incomplete collection, or improper specimen handling were discarded. If creatinine excretion was less than 8% of predicted creatinine excretion, 17 the urine collection was considered incomplete and discarded. Additional 24-hour urine collections were requested from subjects until properly collected samples were obtained. After the volume of each 24-hour collection was recorded, 2 aliquots of 6-mL urine samples were kept for further analysis. One aliquot was acidified and used for determination of citrate level. No acid was added initially to the collecting vessels to avoid interference with urinary titratable acid determination. If urine samples were handled according to our instruction, there was no significant difference in urinary citrate concentrations between samples acidified initially and samples acidified after complete collection (8% 1.1% difference between the 2 techniques). The following urinary constituents were determined: sodium, potassium, chloride, calcium, phosphate, magnesium, citrate, oxalate, urea, creatinine, ammonium, titratable acid, and net gastrointestinal alkaline absorption according to the formula: (Sodium potassium calcium magnesium) (chloride 1.8 phosphate) Electrolyte excretion is in milliequivalents per day, except for phosphate, expressed in millimoles per day with an average valence of 1.8. 18 The average amount of each urine constituent from the 2 urine collections was used for further analysis. Serum samples for determination of electrolyte, creatinine, calcium, and phosphate levels were obtained at the time of urine collection. Control values in this study were obtained from 26 age-matched healthy volunteers, 1 men and 16 women, aged 47.1 2.1 years, with a weight of 61.6 2.1 kg, during the same period and using the same protocol. Dietary Assessment All subjects were interviewed for their diet by a research dietitian. Subjects, following a free-choice diet, performed 3-day dietary records after instruction and verification. Nutrients were calculated by using the INMUCAL (Mahidol University, Bangkok, Thailand) computer program. 19,2 The database for food composition analysis was based on a Thai food composition table developed by Mahidol University. Subsequent Study All subjects were invited to participate in a subsequent study to determine the effect of potassium chloride (KCl) supplement on urinary excretion of citrate. Fractional excretion of citrate (FE-citrate) was determined in this subgroup before starting KCl supplementation. After supplementation for 1 month of 4 meq/d (mmol/d) of KCl in 2 divided doses, an additional two 24-hour urine collections and determination of FE-citrate were performed again, using the same protocol. Subjects were instructed to maintain the same pattern of dietary intake and activity throughout the study. FE-citrate was determined after an overnight fast. In the morning, subjects were hydrated with 1 L of fluid before the beginning of the test. Next, a 2-hour urine collection (8: AM to 1: AM) for determination of creatinine and citrate levels was obtained. During the collection, additional fluid intake (.5 to 1 L) was encouraged to all subjects. Any subject who excreted less than 2 ml of urine during the period was considered to have inadequate hydration, and an

548 DOMRONGKITCHAIPORN, STITCHANTRAKUL, AND KOCHAKARN additional FE-citrate determination was repeated the subsequent day. Serum samples for creatinine and citrate determinations were obtained before and after the 2-hour urine collection. The average of the 2 serum samples was used for FE-citrate calculation according to the formula: 1 (urine citrate serum creatinine)/(serum citrate urine creatinine). The study was approved by the Ramathibodi Hospital Ethics Committee on Human Experimentation. Written informed consent was obtained from all subjects. Analytical Technique Specimens were analyzed for sodium, potassium, chloride, calcium, phosphate, urea, and creatinine content by using autoanalyzer techniques; magnesium, by using an atomic absorption technique; citrate, by using a citrate lyase technique 21 ; oxalate, by using a high-performance liquid chromatographic technique as described previously 22 ; and ammonium and titratable acid, by using a titration technique. 23 All serum and urine samples were analyzed within 1 month and stored at 3 C. Statistical Analysis Data are presented as mean SEM. Correlation between 2 independent variables was determined by means of Pearson correlation. Stepwise backward multiple linear regression analysis was performed to determine the correlation between urinary citrate excretion and known variables that might affect urinary citrate excretion. Input variables for analysis included urine volume and urinary constituents as follows: values for net gastrointestinal alkaline absorption, titratable acid, ammonium, total acid excretion, sodium, potassium, calcium, phosphate, magnesium, oxalate, and urea. Normal distribution of variables used in multiple regression analysis was verified by using the Kolmogorov- Smirnov test. Paired or unpaired Student t-test was used to determine the difference between groups, as appropriate. P less than.5 is considered statistically significant. The SPSS computer program, version 1 (SPSS Inc, Chicago, IL), was used in analyses of data. RESULTS Serum electrolyte, creatinine, calcium, and phosphate levels of subjects are listed in Table 1. No subject had abnormal serum electrolyte, calcium, or phosphate levels. Averages of the two 24-hour urine collections for each urinary constituent are listed in Table 2. Urinary citrate excretion was lower in subjects compared with healthy controls, but the difference was not significant. The distribution of urinary citrate levels among subjects is shown in Fig 1. A total of 5.6% of subjects and 38.2% of healthy controls had urinary citrate levels less than 3 mg/d ( 1.43 mmol/d). Female subjects tended to have greater urinary citrate levels compared with male subjects (377.9 32.3 versus 31. 34.1 mg/d Table 1. Blood Chemistry Test Results for All Recurrent Calcium Stone Formers Blood Chemistry Test Results Sodium (meq/l) 141.2.2 Potassium (meq/l) 4.4.4 Chloride (meq/l) 16.9.2 Bicarbonate (meq/l) 23.8.32 Creatinine (mg/dl).96.3 Calcium (mg/dl) 1.45.97 Phosphate (mg/dl) 3.28.6 NOTE. N 83. To convert sodium, potassium, chloride, and bicarbonate in meq/l to mmol/l, multiply by 1; creatinine in mg/dl to mol/l, multiply by 88.4; calcium in mg/dl to mmol/l, multiply by.2495; phosphate in mg/dl to mmol/l, multiply by.3229. [1.8.15 versus 1.43.16 mmol/d]), but the difference was not statistically significant (P.5). There was no correlation between urinary citrate level and age or body weight among subjects. Urinary excretion of calcium, oxalate, titratable acid, and total acid were significantly greater in subjects than healthy controls, as listed in Table 2. Major food compositions for subjects are listed in Table 3. By using backward stepwise multiple linear regression analysis, values for urinary potassium, net gastrointestinal alkaline absorption, urinary calcium, and titratable acid were independent variables correlated with urinary citrate excretion, as listed in Table 4. Urinary potassium level was the strongest predictor of urinary citrate level. The relationship between urinary potassium and urinary citrate levels is shown in Fig 2. There were significant correlations between titratable acid and urinary excretion of urea (r.6; P.1) and urinary excretion of uric acid (r.5; P.1). To determine the causal relationship between urinary potassium and urinary citrate levels, 4 meq/d (mmol/d) of KCl in 2 divided doses was prescribed to subjects. However, only 58 subjects participated in the subsequent study, defined as the KCl group. Patient characteristics, baseline urinary constituent values, and major food compositions of this subgroup were not different from those of the entire study subjects, as listed in Tables 2 and 3. After 1 month of KCl supplementation, there were significant increases in urinary potassium, chloride, and citrate levels, but no significant changes in levels of other urine constituents were detected, as listed in Table 2.

CAUSES OF HYPOCITRATURIA 549 Table 2. Urinary Constituents for Recurrent Calcium Stone Formers, Healthy Controls, and the KCl Group at Baseline and After Treatment Urinary Constituents All Stone Formers (N 83) Healthy Controls (n 26) Before KCl Group (n 58) After Treatment Volume (L/d) 1.9.9 1.7.11 1.89.11 1.93.11 Creatinine (g/d) 1.15.4 1.15.8 1.15.4 1.11.5 Calcium (mg/d) 193.64 9.96* 151.58 14.82 194.64 12.81 175.46 1.26 Phosphate (g/d).57.2.5.4.56.2.53.2 Magnesium (mg/d) 9.94 3.85 8.85 6.39 85.8 4.5 85.58 3.37 Sodium (meq/d) 171.4 7.2 152.2 12.2 168.68 8.92 165.4 9.15 Potassium (meq/d) 35.7 1.4 34.1 2.6 34.83 1.66 6.19 2.21 Chloride (meq/d) 171.6 6.5 158.8 11.7 167.41 8.5 193.64 8.63 Citrate (mg/d) 337.39 23.83 383.32 43.39 34.15 3. 35.73 27.25 Oxalate (mg/d) 14.41.91 1.67 1.1 14.45 1.12 Urea (g/d) 8.7.28 7.21.52 7.99.33 7.41.3 Uric acid (mg/d) 548.1 16.67 552.7 35.79 554.47 19.6 524.72 2.17 Net gastrointestinal alkaline absorption 19.2 2.6 16.6 3.5 18.38 3.2 15.27 2.79 Ammonium (meq/l) 42.4 2.5 33.7 1.9 42.38 3.6 38.27 3.17 Titratable acid (meq/l) 29.3 1.2* 23.4 1.7 29.34 1.45 28.76 1.2 Total acid excretion (meq/l) 71.7 2.9* 57.1 3.3 71.72 3.57 67.3 3.4 NOTE. To convert creatinine in g/d to mmol/d, multiply by 8.84; calcium in mg/d to mmol/d, multiply by.2495; phosphate in g/d to mmol/d, multiply by 32.29, magnesium in mg/d to mmol/d, multiply by.4114; sodium, potassium, chloride and bicarbonate in meq/d to mmol/d, multiply by 1; citrate in mg/d to mmol/d, divide by 21; oxalate in mg/d to mol/d, multiply by 11.1; urea in g/d to mmol/d, multiply by 35.7. *P.5 compared with the corresponding healthy control value. P.5 compared with the corresponding baseline value in the KCl group. P.1 compared with the corresponding healthy control value. FE-citrate was determined in the KCl group at baseline and after 1 month of KCl supplementation. After KCl supplementation, fasting serum citrate levels were increased significantly, whereas there was no significant alteration in FE-citrate (Table 5). Alterations in urinary potassium and 3 2 Number 1 2-3 1-2 - 1 11-12 1-11 9-1 8-9 7-8 6-7 Fig 1. Distribution of urinary citrate levels in all subjects with recurrent calcium stone formation. 5-6 4-5 3-4 Urinary citrate (mg/d)

55 DOMRONGKITCHAIPORN, STITCHANTRAKUL, AND KOCHAKARN Table 3. Food Composition of All Recurrent Calcium Stone Formers and the KCl Group Food Composition All Subjects (N 83) KCl Group (n 58) Total energy (kcal/d) 1,576.9 47.3 1,56.8 55.9 Carbohydate (g/d) 231.2 7.9 23. 9.6 Fat (g/d) 44.3 1.8 43.3 2.1 Protein (g/d) 6.1 2.4 58.2 2.3 Fruit (g/d) 99.9 1.2 11.3 12.4 Fiber (g/d) 5.7.3 5.6.4 Calcium (mg/d) 44.4 26. 43.5 31.8 urinary citrate levels and FE-citrate for an individual subject after potassium chloride supplement are shown in Fig 3A, B, and C, respectively. To determine the effect of diet on urinary excretion of citrate, total subjects were divided into 4 quartiles based on urinary citrate excretion. Subjects in the highest and lowest quartiles were defined as the high-citrate group (n 21) and low-citrate group (n 21), respectively. Urinary citrate excretion ranged from 474.8 to 1,153.67 (635.49 37.14) mg/d (2.26 to 5.49 [3.3.18] mmol/d]) and 28.58 to 174.21 (114.79 9.2) mg/d (.14 to.83 [.55.4 mmol/d)] for the high- and low-citrate groups, respectively. There was no significant difference in major food compositions between the 2 groups, as listed in Table 6, except for amount of fruit intake: 126.6 17.8, 15.3 17.2, 11.2 22.6, and 64.7 16.1 g/d for the highest to lowest quartile of urinary citrate excretion, respectively. The high-citrate group had significantly greater fruit intake compared with the low-citrate group. DISCUSSION In this study, the incidence of hypocitraturia in recurrent calcium stone formers was high, similar to that reported in our population. 15,16 Although renal tubular acidosis is common in this geographic area, 24-26 it had no contribution to the high incidence of hypocitraturia found in our subjects because subjects with renal tubular acidosis were excluded initially from this study. The unusually low urinary excretion of citrate found in our population, less than half the amounts reported from Western countries, 6 was not limited to only recurrent calcium stone formers. It also was found in healthy controls, although it was less severe. This may indirectly indicate that factors associated with hypocitraturia are also present in the general Thai population. Four independent factors that determined urinary citrate level were identified from our study, including urinary potassium level, net gastrointestinal alkaline absorption, urinary calcium level, and titratable acid. Urinary potassium levels were the strongest variable that determined urinary citrate levels in our subjects. Hypokalemia is a known cause of decreased urinary citrate excretion. 1 Potassium depletion decreases intracellular ph and increases hydrogen secretion into tubular lumen, causing hypocitraturia. 27 However, it should be noted that there was no hypokalemia in our subjects. The relatively low urinary potassium levels might reflect low potassium intake and indicate indirectly that the majority of our subjects might have marginal potassium stores in the body. Fruits are the major source of potassium in the diet. The finding of low urinary potassium levels in our subjects was compatible with the finding of low fruit intake estimated from the dietary record. Subjects in the lowcitrate group had significantly less fruit intake compared with subjects in the high-citrate group. Banana, papaya, mango, and orange were the fruits commonly consumed by our subjects. Potassium contents are approximately 7, 6, 6, and 4 meq/1 g of fresh weight, respectively. The causal relationship between low urinary potassium levels and hypocitraturia also was confirmed by the significant increase in urinary citrate excretion after 1 month of KCl supplementation. However, after potassium supplementation, urinary citrate excretion was still much less than the amount reported from studies in Western populations. 6 Poor compliance or poor absorption of potassium in our subjects was unlikely to Table 4. Independent Variables Affecting Urinary Citrate Excretion in Recurrent Calcium Stone Formers Urinary Constituents 1 2 SE P Potassium 4.19.7.52.1 Net gastrointestinal alkaline absorption 1.33.4.3.1 Calcium.11.44.23.2 Titratable acid 1.7.1.18.9 Constant.23.32.46

CAUSES OF HYPOCITRATURIA 551 12 1 r =.54, p <.1 Urinary citrate (mg/d) 8 6 4 2 Fig 2. Relationship between urinary potassium and urinary citrate levels in recurrent calcium stone formers (r.54; P <.1). 1 2 3 4 5 6 Urinary potassium (meq/d) 7 8 be the cause of the difference. There were few subjects with inappropriate increases in urinary potassium levels after KCl supplementation. If subjects (n 15) with increased urinary excretion after KCl supplementation of less than 2 meq/d (an arbitrary cutoff value for appropriate increase in urinary potassium levels) are excluded, no major change in mean urinary citrate excretion after supplementation is found (378.8 31.5 mg/d [1.8.15 mmol/d]). Therefore, other unidentified factors also may contribute to the low urinary citrate excretion in our population. The increase in urinary citrate levels after potassium supplementation did not result from an increase in tubular excretion of citrate, as indicated by no alteration in FE-citrate after KCl supplementation. However, there was a significant increase in serum citrate levels. This should increase glomerular filtration of citrate and, in turn, enhance urinary citrate excretion. The proportionate increase in serum and urine citrate levels after potassium supplementation suggests that either citrate production or net alkali absorption might have increased between baseline and after potassium supplementation. The finding of a strong association between urinary potassium and urinary citrate levels in our study suggests that determination of urinary potassium excretion should be included as part of the metabolic investigation for recurrent calcium stone formation. Consumption of fruits with high potassium and citrate contents, eg, citrus fruits, would confer a greater potassium and alkaline load to the body. The amount of recommended fruit intake Table 5. Effects of KCl Supplementation on Values for Serum Citrate, Urinary Citrate, and FE-Citrate in the KCl Group and Healthy Controls Baseline After KCl Supplementation Healthy Controls Serum citrate (mg/l) 18.45.55* 2.78.48 2.45.72 Urine citrate (mg/d) 34.15 3. 35.73 27.25 383.32 43.39 FE-citrate (%) 21.9 2.1 19.8 2.7 21.8 2.5 NOTE. To convert serum citrate in mg/l to mmol/l and urine citrate in mg/d to mmol/d, divide by 21. *P.5 compared with healthy controls. P.2 compared with baseline values.

552 DOMRONGKITCHAIPORN, STITCHANTRAKUL, AND KOCHAKARN A 12 B 12 After KCl supplement (meq/d) 11 1 9 8 7 6 5 4 3 r =.6, p <.1 After KCl supplement (mg/d) 1 8 6 4 2 r =.81, p <.1 2 1 2 3 4 5 6 7 8 9 1 11 12 2 4 6 8 1 12 Baseline (meq/d) Baseline (mg/d) C.6.5 r =.28, p <.5 After KCl supplement.4.3.2.1...1.2.3 Baseline.4.5.6 Fig 3. Urinary excretion of (A) potassium, (B) citrate, and (C) FE-citrate before and after KCl supplementation for individual subjects in the KCl group. Straight diagonal lines represent the line of identity. for stone formers should be adjusted individually based on findings in urine composition. Net gastrointestinal alkaline absorption is another factor associated with urinary citrate excretion. 11 This finding conformed with the finding of lower fruit intake in the low-citrate group Table 6. Food Compositions of the High- and Low-Citrate Groups Food Composition High-Citrate Low-Citrate Total energy (kcal/d) 1,569.3 118.2 1,585.7 13.4 Carbohydate (g/d) 224. 15.2 243.8 19.4 Fat (g/d) 46.5 5.6 39.4 2.6 Protein (g/d) 62.7 8.1 57.5 3.8 Fruit (g/d) 126.6 17.8 64.7 16.1* Fiber (g/d) 6.34.82 4.98.65 Calcium (mg/d) 388.1 32.9 31.1 31.4 *P.2 compared with the high-citrate group. because fruits and vegetables are the main sources of alkaline intake to the body. High titratable acid and low urinary calcium excretion also were independent factors associated with hypocitraturia in our study. The formation of complexes between calcium and citrate in urine may decrease citrate transport through the renal tubule, resulting in greater citrate excretion. 28 The cause of low urinary calcium levels in our subjects should result from low dietary calcium intake, supported by the low calcium intake estimated from the dietary record. Dietary assessment for amount of fruit intake and net gastrointestinal alkaline absorption should be added to the metabolic evaluation of patients with recurrent calcium stone formation. It should be noted here that the diet in our subjects, as well as in Thais, is different from the Western diet.

CAUSES OF HYPOCITRATURIA 553 Our subjects tended to consume less protein and calcium compared with Western populations. Our findings are similar to those reported by Hess et al 13 in some aspects. In their study, decreased net gastrointestinal alkaline absorption, low vegetable fiber intake, low urinary calcium levels, low urinary magnesium levels, and low urinary volume were associated with hypocitraturia. Although the causal relationship could not be shown in their study, they also advocated a dietary origin of hypocitraturia in stone formers. In this study, although urinary excretion of calcium, oxalate, and titratable acid were significantly greater in stone formers than healthy controls, they were still within normal limits based on Western populations. However, the contribution of these risk factors to stone formation should not be overlooked for 2 reasons. First, risk factors for urinary tract stone formation are not discreet values with sharp cutoff values between normal and abnormal. Contributions to stone formation of urinary risk factors increase gradually as their urinary concentrations increase. Second, the normal limit in one population may not be the normal limit for another population. 29 In conclusion, hypocitraturia was found as a common risk factor associated with recurrent calcium stone formation and associated with low urinary potassium level, low alkaline absorption, low urinary calcium level, and high titratable acid excretion. Hypocitraturia is predominantly of dietary origin. Estimation of fruit intake should be included in the metabolic evaluation for recurrent calcium stone formation. ACKNOWLEDGMENT The authors thank Professor Amnuay Thithapandha for revision of the manuscript. REFERENCES 1. Meyer JL, Smith LH: Growth of calcium oxalate crystals. II. Inhibition by natural urinary crystal growth inhibitors. Invest Urol 13:36-39, 1975 2. Pak CYC: Citrate and renal calculi. Miner Electrolyte Metab 13:257-266, 1987 3. Nicar MJ, Hill K, Pak CYC: Inhibition by citrate of spontaneous precipitation of calcium oxalate in vitro. J Bone Miner Res 2:215-22, 1987 4. Meyer JL, Smith LH: Growth of calcium oxalate crystal: Inhibition in nephrolithiasis. Urology 21:8-14, 1983 5. Hamm LL: Renal handling of citrate. Kidney Int 38:728-735, 199 6. Parks JH, Ruml LA, Pak CYC: Hypocitraturia, in Coe FL, Favus MJ, Pak CYC, Parks JH, Preminger GM (eds): Kidney Stones: Medical and Surgical Management. Philadelphia, PA, Lippincott-Raven, 1996, pp 95-92 7. Buckalew VM Jr, Canuana RJ: The pathophysiology of distal (type 1) renal tubular acidosis, in Gonick HC, Buckalew VM Jr (eds): Renal Tubular Disorder: Pathophysiology, Diagnosis, and Management. New York, NY, Dekker, 1985, pp 357-386 8. Rudman D, Dedpnis JL, Foutain MT, et al: Hypocitraturia in patients with gastrointestinal malabsorption. N Engl J Med 33:657-661, 198 9. Simpson DP: Citrate excretion: A window on renal metabolism. Am J Physiol 244:F223-F234, 1983 1. Fourman P, Robinson JR: Diminished urinary excretion of citrate during deficiencies of potassium in man. Lancet 2:656-657, 1953 11. Pak CYC: Citrate and renal calculi: New insights and future directions. Am J Kidney Dis 17:42-425, 1991 12. Hodgkinson A: The relation between citric acid and calcium metabolism with particular reference to hyperparathyroidism and idiopathic hypercalciuria. Clin Sci 24:167-178, 1963 13. Hess B, Michel R, Takkinen R, Ackermann D, Jaeger PH: Risk factors for low urinary citrate in calcium nephrolithiasis: Low vegetable fiber and low urine volume to be added to the list. Nephrol Dial Transplant 9:642-649, 1994 14. Kok DJ, Iestra JA, Doorenbos CJ, Papapoulos SE: The effects of dietary excesses in animal protein and in sodium on the composition and the crytallization kinetics of calcium oxalate monohydrate in urines of healthy men. J Clin Endocrinol Metab 71:861-867, 199 15. Nimmannit S, Malasit P, Susaengrat W, Ong-Aj- Yooth S, Vasuvattakul S, Pidetcha P: Prevalence of endemic distal renal tubular acidosis and renal stone in the northeast of Thailand. Nephron 72:64-61, 1996 16. Usui Y, Matsuzaki S, Matsushita K, Shima M: Urinary citrate in kidney stone disease. Tokai J Exp Clin Med 28:65-7, 23 17. Fournier A, Achard JM: Mnemotechnical note on the use of Cockcroft creatinine clearance formula for the validation of a 24-h urine collection. Nephrol Dial Transplant 15:1677-1678, 2 18. Oh MS: A new method for estimation of GI absorption of alkaline. Kidney Int 36:915-917, 1989 19. Institute of Nutrition: Food Composition Database for INMUCAL Program. Salaya, Nakhonpathom, Thailand, Institute of Nutrition, Mahidol University, 1997 2. Matsuda-Inoguchi N, Shimbo S, Zhang Z-W, et al: Nutrient intake of working woman in Bangkok, Thailand, as studied by total food duplicate method. Eur J Clin Nutr 54:187-194, 2 21. Toftegaard Nielsen T: A method for enzymatic determination of citrate in serum and urine. Scand J Clin Lab Invest 36:513-519, 1976 22. Hagen L, Walker VR, Sutton RAL: Plasma and urinary oxalate and glycolate in healthy subjects. Clin Chem 39:134-138, 1993 23. Chan JCM: The rapid determination of urinary titratable acid and ammonium and evaluation of freezing as a method of preservation. Clin Biochem 5:94-98, 1972

554 DOMRONGKITCHAIPORN, STITCHANTRAKUL, AND KOCHAKARN 24. Nilwarangkur S, Nimmannit S, Chaovakul V, et al: Endemic primary distal renal tubular acidosis in Thailand. Q J Med 74:289-31, 199 25. Domrongkitchaiporn S, Pongsakul C, Stitchantrakul W, et al: Bone mineral density and histology in distal renal tubular acidosis. Kidney Int 59:186-193, 21 26. Domrongkitchaiporn S, Khositseth S, Stitchantrakul W, Tapaneya-olarn W, Radinahamed P: Dosage of potassium citrate in the correction of urinary abnormalities in pediatric distal renal tubular acidosis patients. Am J Kidney Dis 39:383-391, 22 27. Adam WR, Koretsky AP, Weiner MW: 31 P-NMR in vivo measurement of renal intracellular ph: Effects of acidosis and K depletion in rats. Am J Physiol 251:F94-F91, 1986 28. Walser M: Ion association. VI. Interactions between calcium, magnesium, inorganic phosphate, citrate and protein in normal human plasma. J Clin Invest 4:723-73, 1961 29. Marshall VR, Ryal RL. An evidence-based approach to hypercalciuria: Is it really necessary for the study of the role of calcium in urolithiasis? Curr Opin Urol 8:327-33, 1998