Correspondence between stone composition and urine supersaturation in nephrolithiasis

Kidney International, Vol. 51 (1997), pp. 894 900 Correspondence between stone composition and urine supersaturation in nephrolithiasis JOAN H. PARKS, MARK COWARD, and REDRIC L. COE Program in Nephrology, Department of Medicine, University of Chicago, Chicago, Illinois, USA Correspondence between stone composition and urine supersaturation in nephrolithiasis. Supersaturation (SS) with respect to calcium oxalate monohydrate (COM), brushite (Br) and uric acid (UA), obtained in three 24-hour pretreatment urine samples from patients with stone disease were compared to the mineral composition of stones passed by the same patients to determine whether sparse urine SS measurements accurately reflect the long-term average SS values in the kidney and final urine. Among males and females elevation of SS above same sex normals corresponded to composition. As well, treatments that reduced stone rates also reduced these SS values. The degree of calcium phosphate (CaP) admixture was accurately matched by shifting magnitudes of COM and Br SS. As well, increasing CaP content was associated with falling urine citrate and rising urine ph, suggesting renal tubular acidosis. We conclude that sparse urine SS measurements accurately track stone admixtures, and are a reliable index of average renal and urine SS. Supersaturation (SS), the presence of a material in solution at a concentration above its own solubility [1], is the driving force for crystal formation [2]. Relevant SS for the majority of human stones include calcium oxalate monohydrate (COM), calcium phosphate phases such as brushite (Br) and octocalcium phosphate, uric acid (UA), and struvite. We neglected struvite in this study, as it concerns only stones of infectious origin [31, and considered Br as the main CaP phase [4]. So-called 'calcium' stones are usually made predominantly of CaOx [5], but a majority also contain CaP, mainly as apatite. COM, Br, and UA SS values can be calculated in human urine samples using measurements of pertinent ligands [6], and directly measured by adding crystals to urine aliquots and determining the fall in relevant ion concentrations [7]. We [8 12] and others have presented such measurements, and generally have found elevated SS values in stone forming patients compared to normal people. One epidemiological study found SS to he among the principal factors that discriminate between stone forming and non-stone forming members of a study group [13]. An open question with regard to SS measurements is how well a sparse sample, of perhaps three pretreatment 24-hour urine collections, can represent the average SS present over the months to years during which stones form and grow. Diet, habits, change in work all would be expected to alter SS averages over the Received for publication July 3, 1996 and in revised form September 5, 1996 Accepted for publication September 5, 1996 1997 by the International Society of Nephrology 894 considerable time scale of stone production. Possibly, urine SS fails to reflect important SS values in the nephron because of heterogeneity. or example, we have presented evidence for a constant calcium phosphate SS in the thin segment of Henle's loop [14] that has no necessary counterpart in urine SS. actors in the nephron, such as nucleating sites [15 18], and inhibitors or promoters [19, 20] affect crystallization. These may so overweigh SS effects that variations of urine SS are not well linked to stone composition. inally, Br SS may not fully reflect forces for production of phosphate phases in stones, which are mainly apatite. ortunately, the stones themselves offer a strategy for assessing this question. The stone composition should reflect the SS averages during the creation and growth of the stone, so the stone is a kind of geological artifact of SS conditions. If sparse urine samples are representative, their measured SS values should parallel the phase compositions of the stones actually formed and passed by the patients who provide those urine samples. If such a parallel cannot be found, the hypothesis that sparse samples reflect average nephron conditions should be rejected. We present here the comparison of urine SS and stone admixtures among a large group of patients, as a critical test of the sparse urine sample hypothesis. Methods Patients and normal subjects rom our total of 1085 patients with crystallographic analyses of stones, we selected those with no systemic cause of stones such as hyperparathyroidism, enteric disease, laxative abuse, or cystinuria. We also excluded patients whose stones contained any struvite or any unusual material, such as sodium, potassium or ammonium urate, or crystallized drug. This left 585 patients. Each had been evaluated prior to treatment with SS measurements, and had formed stones that fell into one of five discrete groups: (1) CaOx stones, defined as > 80% calcium oxalate in the average of all stones, less than 20% calcium phosphate in the average of all such stones that contain any calcium phosphate, and no uric acid in any stone (316 males and 85 females); (2) CaP stones, > 50% calcium phosphate average for all stones, no uric acid, < 50% calcium oxalate for the average of all stones that contained calcium oxalate (19 males and 24 females); (3) UA stones, all stones contained only uric acid (11 males and 3 females); (4) mixed (calcium oxalate uric acid) stones, both crystals in stones, in any amounts (67 males and 12 females); (5) CaP/CaOx stones, calcium oxalate content above 0 and calcium phosphate

Parks et al: Supersaturation and stone composition 895 Table 1. Urine supersaturations in stone forming patients and controls N COM SS Br SS UA SS - Type Sex C RX C RX C RX C RX Normal Men 65 8.4 0.52 1.71 0.17 1.33 0.14 Women 54 6.66 0.45 1.12 0.16 1.01 0.13 CAOX Men 935 1934 9.28 0.16 5.75 Oöfl 1.62 0.04 1.27 0.02"' 1.32 0.04 0.61 0.02' Women 254 479 0.34" 5.27 15" 1.60 0.08" 1.10 004" 1.43 0.08" 0.47 0.03"' CAP Men 58 102 8.80 0.57 5.36 0.34" 2.41 0.13" 1.81 0.09" I 0.51 0.09" 0.11 0.01' Women 69 154 7.68 0.52 4.45 0.27"' 1.96 0.14" 1.45 0.09 0.55 0.08" 0.18 002ad Uric acid Men 33 58 5.02 0.54" 4.52 0.36a 0.45 0.08 0.65 0.09" 1.84 0.24" 1.14 0.141 Women 9 17 8.02 2.3 3.60 0.40 0.622 0.09 1.12 0.16 2.03 0.48" 0.45 0.14' Mixed Men 190 390 8.68 0.34 5.97 0.17"' 0.73 0.06" 1.00 0.05"' 2.34 0.09" 1.13 0.06" Women 36 88 9.22 0.70' 6.42 0.42" 1.02 0.16 1.12 0.09 1.79 0.18" 0.77 0.08J Values are SS SEM. Type is type of stones formed (Methods) Abbreviations are: N, number of samples provided by subjects in the group specified; C, pretreatment, RX, during treatment samples; COM, Br and UA are calcium oxalate, calcium phosphate, and uric acid supersaturations, respectively (Methods); UA SS in CaP stone formers, COM SS in male VA stone formers, Br SS in UA and Mixed stone formers all are significantly below corresponding same sex controls. Differs from same sex normal <, " P < 0.01, "P < 0.05 "Differs from pretreatment, same sex, P <.001, "P < 0.01 content > 20% and < 50%, no uric acid (31 males and 19 females). Our control group was comprised of 67 non-stone forming people (33 males and 34 females). All patients were studied using a protocol based on three 24-hour urines with three corresponding blood samples drawn 12 hours after the last meal [9]. Stone prevention was also based upon a protocol detailed elsewhere [21, 22], which aimed at reversing the main nonsystemic causes of stones: idiopathic hypercalciuria, hypocitraturia, hyperoxaluria, hyperuricosuria, and low urine volume and ph. During treatment, 24 hour urines and corresponding blood samples were obtained at about six weeks, and yearly thereafter, to monitor treatment. Because this facility is a referral center, virtually all patients were active stone formers at the time of their initial evaluation. In general, referral was created by a recent stone or procedure, and stone rates were often higher near to the time of referral than in the past history of the patients. Stones were analyzed by routine commercial laboratories. We added the % values for calcium oxalate monohydrate and dihydrate to obtain the % CaOx value for a stone, and added all phases of calcium phosphate together, in the same manner, to obtain the %CaP. Phases included apatite, brushite, and octocalcium phosphate. or uric acid, we took the % as noted. Laboratoiy measurements and calculations of SS In urine we measured calcium, phosphate, creatinine, uric acid, magnesium, sodium, potassium, ph, volume, citrate, oxalate, sulfate, chloride, and ammonium ion. rom this set of urine measurements SS was calculated with respect to calcium oxalate monohydrate, brushite, and undissociated uric acid using the software program EQUIL [6]. Statistical methods SS values among patients and normals were compared using contrasts in the general linear model [23]. Essentially, this is a multivariate analysis of variance. Some comparisons were done with urine volume as the covariate, to permit assessing differences in SS holding volume to the joint mean of comparison groups. In this, SS was the dependent variable, volume the independent variable, and the group as defined by stone type was the grouping variable. or presentation, we calculated means, and the standard error (sem) for all SS values by sex and stone type. or all analyses all SS values were transformed by square root to achieve normality. The distribution of the square roots of a set of numbers is often more normal in its distribution than the parent distribution, as was the case here. We performed the contrasts two ways: sample based, using all 24-hour urine samples from each group and normals, and subject based, using either pretreatment or treatment means from each patient. Sample based analysis weighted individuals by their numbers of samples, whereas subject based means that each individual contributed equally to the subject based means. Volume adjustment was applied only in comparisons of patient pretreatment results to sex-matched normal subjects; all treatment comparisons to either normals or pretreatment used no covariates because treatment includes a deliberate increase of urine volume, to reduce SS, whereas pretreatment values reflect the actual ambient hydration during active stone formation. We performed our analyses by first excluding Group 5, as defined above. This was to achieve a sense of SS correspondence with stone phases at the extremes of composition. We then performed a separate and more detailed analysis of CaP admixture, in which Group 5 was considered along with Groups 1 and 2. or this, Groups 3 and 4 were omitted, because uric acid forms at relatively low urine ph values, and uric acid stones have little if any CaP phases in them. Results Sample based analysis, Groups I to 4 Our main hypothesis is that significant SS elevations should occur mainly in the boxed regions of Table 1, whose cells represent the matching of SS and crystal composition. The prediction is, in part, fulfilled. Pretreatment SS is high for women in all cells (lower rows in boxed regions). Men were the same, except for COM SS that was not above normal in patients with either CaOx stones or mixed stones (upper rows in boxed areas). There were no significant elevations of SS outside the boxes, except among women with CaOx stones who had high Br and Ua SS. All other significant departures of SS from normal were in a

896 Parks et al: Supersaturation and stone composition Cl) Cl) 12.25 9.00 I 6.25 12.96 ii.56 'I) g.0. 7.6 6.7 2 76 4.84 400 A c- ig. 1. A. Male stone formers plotted on the three SS axes. Actual SS values and P values for comparisons are in Table 1. or the CaPICaOx group values and comparisons are in the text (Results). The groups represented by the descriptive labels are: CaOx, Group 1; CaP, Group 2; UA, Group 3; Mixed, Group 4; CaP/CaOx, Group 5. B. emale stone formers plotted on the three SS axes. Actual SS values and P values for comparisons are in Table 1. or the CaP/CaOx group values and comparisons are in the text (Results). Labels and groups as in igure 1A. downward direction, as one would expect from the mismatch of SS to stone crystal composition (Table 1). If visualized along the three SS coordinates of Table I (ig. 1), stone formers segregated to the SS regions appropriate to their stone composition group. The remarkable difference in SS between normal women and normal men (P < ) is best seen by comparing igure 1 a and b. Urine volumes of men with CaOx and CaP stones exceeded the volumes of normal men (Table 2), whereas for women the opposite was true as urine volumes were below normal for those with CaOx, Ua and mixed stones. An SS that is above normal in relation to urine volume would lead to a corresponding elevation of SS above normal whenever urine volume passes through normal. SS comparisons of patients and normals using ANOVA with volume as covariate (Table 2) were the same as for unadjusted SS, except that men and women alike displayed high SS values that corresponded to their stone compositions (boxed areas). The exceptions were patients with CaP stones who had elevated COM SS elevations, men with mixed stones who did not have elevated COM SS, and the three women with UA stones whose Ua SS elevation did not reach significance. Treatments reduced stone relapse among those patients with follow-up data. Relapse occurred in 42 males with calcium oxalate stones, 3 with CaP stones, none with uric acid stones, and 8 with mixed stones, or 17%, 20%, 0% and 15%, respectively (total numbers of patients in Methods). Among women, corresponding values were 9, 5, 0, and 0, or 13%, 26%, 0%, and 0%, respectively. Overall, only 16% of men and 14% of women relapsed. These results have been detailed elsewhere, and occurred over long periods of treatment 24, 25]. One would expect that such effective treatment would act through reduced SS, and such a reduction was very evident in all groups (Table 1 and ig. 1). Relevant SS was reduced to or below normal (boxed areas), except for Br SS among women with CaP stones. As well, values of SS were reduced compared to pretreatment (Table 1 and ig. 1). In other words, treatment reduced SS greatly, in relation to both normal and pretreatment values, and in the direction required to reduce formation of the relevant solid phases for each group. Subject based analysis, Groups 1 to 4 Because patients provided three samples pretreatment versus one or two for each normal person, and because during treatment patients provided widely differing numbers of samples, we represented each patient by the mean of all pretreatment and treatment values, and each normal person by the average of all samples. Our main hypothesis, that significant SS elevations will occur mainly in the boxed areas of Table 3, was still supported except for the mixed stone formers. In fact, there were no significant SS elevations outside those areas except for COM SS in CaP stone formers. Treatment reduced SS corresponding to stone type in all instances except male CAP stone formers (Table 4). Other reductions of SS unrelated to stone type also occurred, but this neither supports nor opposes our basic hypothesis; effective treatments reduced relevant SS values. The correspondence of subject and sample centered contrasts, especially in treatment comparisons, shows that patients with large numbers of samples did not importantly bias the sample based comparisons. Admixture and SS, Groups 1, 2 and 5 Our analysis has thus far relied on exclusion of the middle, with respect to calcium phosphate admixturcs with calcium oxalate. However, an implicit prediction of our main hypothesis is that the excluded middle group of patients (Group 5, Methods) should show triple coordinates for SS between the locations of Groups 1 and 2. This middle group, with stones that contain between 20 and 50% calcium phosphate, should show SS values midway between those of the predominant calcium oxalate stone formers and those patients whose stones are above 50% calcium phosphate. This is in fact the case (ig. 1 a, b). The critical COM and Br SS values (ig. 2) show a close correspondence with increasing CaP admixture among women, and a similar pattern among men but compressed in scale because the dynamic range of SS was smaller. Mean 24-hour urine ph values rose progressively with CaP admixture, being slightly below normal among Group I patients,

Sex Volume Patient Parks et al: Supersaturation and stone composition 897 Table 2. Volume adjusted values of supersaturation CoM SS Br SS UA SS Normal Patient Normal Patient Normal Patient CAOX M 1.64 0.024" 7.54 0.52 9.33 0.14" 1.53 0.14 1.64 0.04 1.13 0.13 1.34 0.034 W 1.27 0.04" 7.56 0.60 10.15 0.2r' 1.26 0.17 1.57 0.08 1.22 0.15 1.38 0.07 CAP M 1.855 0.09" 7.83 0.51 947 0.54 1.56 0.14 2.58 0.15" 1.19 0.11 0.664 0.12 W 1.75 0.1 6.39 0.46 7.884 0.41 1.07 0.15 1.99 0.13" 0.96 0.1 0.594 0.08 UA M 1.77 0.17 8.13 0.46 5.58 0.62" 1.67 0.14 0.52 0.18" 1.2 0.14 2.06 0.19 W 1.08 0.14' 6.81 0.51 7.16 1.26 1.169 0.13 0.315 0.33 1.05 0.13 1.74 0.32 Mixed M 1.39 0.034 8.28 0.53 8.68 0.311 1.69 0.12 0.729 0.069 1.28 0.13 2.37 0.078a W 1.25 0.086c 0.42 8.63 0.52 1.22 0.13 0.87 0.16 1.09 0.13 1.67 0.15 Urine volumes for normal men and women, respectively are 1.33 0.49 and 1.56 0.10 liter/day. Values are volume adjusted SEM, except volumes, which are actual values. Abbreviations are in Table 1, except that M and W are men and women, respectively. "Differs from normal, same sex, P < ; b P < 0.01; "P < 0.05 Table 3. Subject centered SS comparisons between patients and normals UA Stone Sex N COM SS Br SS SS CAOX M 316 85 CAP M 19 24 UA M 11 3 Mixed M 67 12 IO.oolb I 0.12 0.03" o.ooi"i 0.07 0.22 0.028" 0001" I 0.012" 0.024" " 0.o50 b >< 0.009" o.oola 10.004 o 0.64 0.22 0.20 E [ö1 o.oola 0001b 0.43 003h j C) P values are from contrasts in the linear model (Methods). All high SS values are in boxed areas of corresponding SS and stone type except for COM SS in CAP Stone forming women. "Below normal SS b Above normal SS C,) C,) 11 10 9 8 7 CaOX M CaOX i M normal normal CaP/CaOX M CaP/CaOX M CaPt CaP Table 4. Subject centered SS comparisons between untreated and treated patients Stone Sex N COM SS CAOX M 243 68 CAP M 15 UA 19 M 10 2 Mixed M 53 11 0.025 0.79 0.22 0.003 Br SS 0.006 0.002 0.119 0.209 0.297 0.152 0.392 UA SS 0.716 0.779 0.027 0.026 0.024 P values from contrasts in linear model (Methods); all significant changes were in a downward direction. SS values in all boxed areas of corresponding SS and Stone type fell except fur Br SS in men CAP stone formers; other reductions of SS also occurred as noted. and above normal in Groups 2 and 5 (ig. 3). Urine citrate concentration fell as ph rose (ig. 3). All differences between groups by sex were significant except as noted in the legend to igure 3. In other words, at least these two critical determinants of Br SS, ph and citrate, varied in a direction to raise it in patients whose stones were enriched with CaP (ig. 3). The other main determinants of Br SS, calcium, oxalate, and phosphate excretions and concentrations did not differ between the relevant groups under discussion except for higher calcium excretion among women in Groups 2 and 5 compared to women in Group 1 (not 6 1.0 1.5 2.0 2.5 SS brushite ig. 2. Male and female patients plotted along their CaOx and CaP supersaturation axes. Stone groups are defined in the Methods section. Labels and groups are as in igure 1. COM SS Groups 1, 2, and 5 versus normal, P values were all not significant except for females in Groups 1 and 5 (P < ). or Br SS Groups 1, 2, and 5 versus normal, P values were all < 0.01 except for men in Group 1 (NS). shown). Urine volumes in Group 5 did not differ from normal or the other groups, except for women (Group 5, 1.38 liter/day vs. Group 2, Table 2; P < 0.05). Discussion Our findings strongly support the hypothesis that urine COM, Br, and UA SS, as measured in three 24-hour outpatient urine collections, closely reflect the average driving forces in the nephron that control formation of the solid phases of kidney stones. Even though stones themselves are widely spaced in time, and probably grow over months to even years, SS in our three urine samples closely matchs the admixtures of CaOx, CaP and UA in the stones. Since stones should reflect the integrated time average of SS values in tubule fluid and urine, this close matching

898 Parks et al: Supersaturation and stone composition 3.0 have high COM SS values compared to normal women. Normal male COM SS values are about those of male patients, whatever the CaP or UA admixture of their stones. Normal female COM SS 0 normal values, however, are remarkably lower than among patients or normal males. This difference may well be the primary reason that 2.5 0 nearly 80% of stone formers are men [24]. In part, the lack of a M normal clear male patient COM SS elevation above male normals is CaOX because males with CaOx stones have higher urine volumes than E normal men. When volume is used as a covariate, COM SS of 1) male stone formers is decisively above normal male SS, meaning cci 2.0 M CaOX 0 0 CaPICaOX that variations of urine volume through the normal range, from ci) C ordinary effects of diet, habit, and environment, pose a greater stone risk. In a well known study [44], high SS values for CaOx and D A M CaP/CaOX CaP were clearly demonstrated in patients with stones. Of interest, the control group was half female whereas the stone groups 1.5 were about 70% male. a The high urine volume of male stone formers is not of an M CaP obvious origin. Perhaps men, but not women, are given, and heed, I 1.0 A advice to drink more water as a stone prevention. This seems a bit 5.90 6.02 6.14 6.26 6.38 6.50 unlikely. Men with mixed CaOx/UA stones do not have higher volumes, as women do not. Our patients are referred here because Urine ph of active stone disease and therefore the initial evaluation data, ig. 3. Relationship between urine citrate concentration and ph. Male and which show the high male urine volumes, are taken from active female stone groups are defined in the Methods section. Symbols are: stone formers. Perhaps hypercalciuria or some other mechanism males, triangles; females, circles; open are normals; light gray, the CaOx group; darker gray, the CaP/CaOx group; and solid symbols are CaP. CaP reduces the urine concentrating power among men. patients differed from all other groups, by sex, in both citrate and ph, P < Our treatment protocol results in a low frequency of relapse 0.01 all comparisons. CaP/CaOx differed from CaOx and normal except (Results). One would expect that SS should fall with such for males versus normals in ph, and females versus CaOx in citrate and treatments to at least the levels seen in normals. Among men and ph. CaOx differed from normal only in citrate, P < 0.05 and <, females and males, respectively. Labels and groups are in igure 1. women, SS fell significantly compared to pretreatment. SS did not always fall below normal, but was no higher than normal except among women with CaP stones (Table 1). This fact, coupled with the fall in stones, is consistent with the idea that SS is a main but validates the use of urine samples as a strong indicator of the not sole factor in stone production. Inhibitors, promoters, and average conditions that produce stones. local nucleation and adhesion sites are well known and much In the case of CaP admixture, our results demonstrate a studied additional factors [45 48]. surprising degree of predictive value for SS. When patients were Prior studies from this laboratory are fully compatible with the divided into Groups 1, 2 and 5, representing variable degrees of present one. Using crystal addition, we [8] found higher COM SS CaP admixture, we found close correspondence between the among patients who had formed at least one calcium stone, balance of COM SS and Br SS and stone admixture, especially without uric acid. SS fell with treatment. Among patients with at among women. The pattern among men was the same but least one calcium stone and hyperuricosuria, we [10] found high miniaturized in dynamic scale along the COM SS axis because SS with respect to sodium hydrogen urate and uric acid, which normal men have high COM SS compared to normal women. We suggested a role for uric acid crystallization in calcium stone were able to find at least part of the mechanisms for high Br SS in production. Among patients with over 10 calcium stones per a surprising inverse correlation between urine ph and citrate. person, we failed to find abnormal COM SS, perhaps because Elsewhere, we [26] have shown that urine citrate rises greatly with such patients had only modest hypercalciuria [49]. Using crystal alkali, in agreement with others [27 33]. In pregnancy [26], we addition, we have in the past found higher COM SS among found a rise of urine citrate with urine ph, though less marked normal men than normal women, and a corresponding elevation than among nonpregnant women [26]. The renal basis for the of SS among women stone formers compared to normal women, inverse correlation shown here is unknown. However, it suggests with a lack of such an elevation among men with stones, who also common hut mild renal tubular acidosis among patients with even were found to have a higher than normal urine volume [9]. Given modest CaP admixtures. the many years that separate our two studies, the correspondence Uric acid admixtures also showed an excellent correlation of findings is impressive to us. We have also found high COM and between SS and composition. However, mixed CaOxJUA stone Br SS in pregnancy resulting from pregnancy hypercalciuria, formers of both sexes do not have higher than normal COM SS. increased urine ph, and a failure of urine citrate to rise in a Possibly, the CaOx component of their stones is due to an manner appropriate to rising urine ph [26]. We [50] have found interaction between uric acid crystallization and CaOx crystallization, such as has been explored in the hyperuricosuria calcium both COM and UA SS, whereas patients who formed pure UA that patients with mixed CaOx/UA stones exhibit elevation of oxalate nephrolithiasis complex [15, 24, 34 43]. This could permit stones had more marked UA SS. COM to develop even without a remarkable increase of COM SS. Large numbers of studies by other laboratories have shown high We have found, in the course of this work, that normal men SS values in stone formers, with a reduction in COM and Br SS by

Parks et al: Supersaturation and stone composition 899 treatments that include water [121, medications [11, 40, 51 60], and altered diet sodium or acid [61, 62]. Simplified approximations to calculated SS have been extensively validated and show a pattern of higher values among patients than normal subjects [1, 63]. inally, in a population study, Thun and Schober [13] found that COM SS values strongly correlated with the risk of stone formation, with family history being the other main correlate. All of these are consistent with this study, which adds to them the correspondences between measured SS and stone compositions, and the important sex differences not always clear in other studies. These new additions represent a critical test of the hypothesis that SS in sparse urine samples accurately represent SS in the kidney and urine during the time period needed for formation of human kidney stones. Being a positive result, the study supports the hypothesis. Naturally, other critical tests of this hypothesis should be considered and performed, as the hypothesis is central to the practice of stone prevention at this time. Acknowledgment This study was supported by NIH NIDDDK DK47631. Reprint requests to redric L. Coe, M.D., Nephrology Program, University of Chicago, MC5100, 5841 S Maryland Avenue, Chicago, Illinois 60637, USA. References 1. TISELIUS H: Solution chemistry of supersaturation, in Kidney Stones: Medical and Surgical Management, edited by COE L, AVUS MJ, PAK CYC, PARKS JH, PREMINGER GM, Philadelphia, Lippincott-Raven, 1996, p33 2. INLAYSON B: Physicochemical aspects of urolithiasis. Kidney mt 13:344 360, 1978 3. GRIITH DP, BRUCE RR: Infection (urease)-induced stones, in Nephrolithiasis, Contemporary Issues in Nephrology, edited by C0E L, BRENNER BM, STEIN JH, New York, Churchill Livingstone, 1980, p 231 4. PAK CYC, EANES ED, RUSKIN B: Spontaneous precipitation of brushite in urine: Evidence that brushite is the nidus of renal stones originating as calcium phosphate. Proc Nat Acad Sci USA 68/7:1456 4564, 1971 5. HERRING LC: Observations on the analysis of ten thousand urinary calculus. J Urol 88:545 555, 1962 6. INLAYSON B: Calcium stones: Some physical and clinical aspects, in Calcium Metabolism in Renal ailure and Nephrolithiasis, edited by DAVID DS, New York, John Wiley & Sons, 1977, p 337 7. PAK CYC, HAYASHI Y, INLAYSON B, CHU S: Estimation of the state of supersaturation of brushite and calcium oxalate in urine: A comparison of three methods. J Lab Clin Med 89:891 909, 1977 8. WEBER DV, COE L, PARKS JH, DUNN MS, TEMBE V: Urinary saturation measurements in calcium nephrolithiasis. Ann Intern Med 90:180 184, 1979 9. PARKS JH, COE L: A urinary calcium-citrate index for the evaluation of nephrolithiasis. Kidney mt 30:85 90, 1986 10, Cou L, STRAUSS AL, TEMBE V, L DUN S: Uric acid saturation in calcium nephrolithiasis. Kidney mt 17:662 668, 1980 11. PAK CY, ULLER C: Idiopathic hypocitraturic calcium-oxalate nephrolithiasis successfully treated with potassium citrate. Ann Intern Med 104:33 37, 1986 12. PAK CYC, SAKIIAEE K, CROWTHER C, BRINKLEY L: Evidence justifying a high fluid intake in treatment of nephrohthiaisis. Ann Intern Med 93:36 39, 1980 13. THUN MJ, SCIIOBER 5: Urolithiasis in Tennessee: An occupational window into a regional problem. Am J Public Health 81 :587 591, 1991 14. ASPLIN JR, MANDEL NS, COE L: Evidence for calcium phosphate supersaturation in the loop of Henle. Am J Physiol 270:604 613, 1996 15. DEGANELL0 5, CH0U C: The uric acid-whewellite association in human kidney stones. Scan Electron Microsc:927 933, 1984 16. LIESKE JC, WALSH-REITZ MM, TOBACK G: Calcium oxalate monohydrate crystals are endocytosed by renal epithelial cells and induce proliferation. Am J Physiol 262:622 630, 1992 17. LIESKE IC, TOBACK G: Regulation of renal epithelial cell endocytosis of calcium oxalate monohydrate crystals. Am J Physiol 264:800 807, 1993 18. LIESKE IC, LEONARD R, SWIr H, TOBACK G: Adhesion of calcium oxalate monohydrate crystals to anionic Sites on the surface of renal epithelial cells. Am J Physiol 39:192 199, 1996 19. ASPLIN J, DEGMtELLO 5, NAKAGAWA YN, COE L: Evidence that nephrocalcin and urine inhibit nucleation of calcium oxalate monohydrate crystals. Am J Physiol 261 :824 830, 1991 20. C0E L, PARKS JH, NAKAGAWA Y: Inhibitors and promoters of calcium oxalate crystallization: Their relationship to the pathogenesis and treatment of nephrolithiasis, in Disorders of Bone and Mineral Metabolism, edited by C0E L, AVUS MJ, New York, Raven Press. 1992, p 757 21. C0E L, PARKS JH, ASPLIN JR: The pathogenesis and treatment of kidney stones Medical progress. N EngI J Med 327:1141 1152, 1992 22. COE L, PARKS JH: Nephrolithiasis: Pathogenesis and Treatment (2nd ed). Chicago, Year Book Medical Publishers, 1988, p 1 23. WILKINSON L: SYSTAT: The System for Statistics. Evanston, Systat, Inc., 1990, p 1 24. COE L: Treated and untreated recurrent calcium nephrolithiasis ir patients with idiopathic hypercalciuria, hyperuricosuria, or no metabolic disorder. Ann Intern Med 87:404 410, 1977 25. PARKS JH, COE L: An increasing number of calcium oxalate stone events worsens treatment outcome. Kidney Int 45:1722 1730, 1994 26. MAIKRANZ P, COE L, PARKS I, LINDHEIMER MD: Nephrolithiasis in pregnancy. Am J Kidney Dis 9:354 358, 1987 27. PAK CY, SAKHAEE K, ULLER Cl: Physiological and physiochemical correction and prevention of calcium stone formation by potassium citrate therapy. Trans Assoc Am Phys 96:294 305, 1983 28. PAK CYC: Citrate and renal calculi. Miner Electrol Metab 13:257 266, 1987 29. PAK CYC, PETERSON R, SAKHAEE K, ULLER C, PREMINGER GM, REISCH J: Correction of hypocitraturia and prevention of stone formation by combined thiazide and potassium citrate therapy in thiazide-unresponsive hypercalciuric nephrolithiasis. Am J Med 79: 284 288, 1985 30. SAKHAEE K, ALPERN R, JACOBSON HR. PAR CYC: Contrasting effects of various potassium salts on renal citrate excretion. J Clin Endocrinol Metab 72:396 400, 1991 31. NICAR MJ, SKURLA C, SAKHAEE K, PAK CYC: Low urinary citrate excretion in nephrolithiasis. Urol 21:8 13, 1983 32. SAKHAEE K, WILLIAMS RH, OH MS, PADALINO P, ADAMS-HUET B, WHIT50N P, PAK CY: Alkali absorption and citrate excretion in calcium nephrolithiasis. J Bone Miner Res 8:789 794, 1993 33. HAMM LL, ALPERN RJ: Regulation of acid-base balance, citrate, and urine ph, in Kidney Stones: Medical and Surgical Management, edited by COE L, AVUS MJ, PAR CYC, PARKS JH, PREMINGER GM, Philadelphia, Lippineott-Raven, 1996, p 289 34. DEGANELLO 5, CHOU C: The uric acid-whewellite association in human kidney stones. Scanning Microsc 4:1545 1550, 1985 35. C0E L: Hyperuricosuric calcium oxalate nephrolithiasis. Adv Exp Med Biol 128:439 450, 1980 36. COE L: Hyperuricosuric calcium oxalate nephrolithiasis. Kidney Int 13:418 426, 1978 37. COE L, MORAN E, KAVALICI AG: The contribution of dietary purine over-consumption to hyperpuricosuria in calcium oxalate stone formems. J Chronic Dis 29:793 800, 1976 38. PAK CY, ARNOLD LH: Heterogeneous nucleation of calcium oxalate by seeds of monosodium urate. Proc Soc Exp Biol Med 149:930 932, 1975 39. PAK CYC, WATERS 0, ARNOL[) I, HOLT K, Cox C, BARILLA DE: Mechanism for calcium urolithiasis among patients with hyperuricosuria. J Clin Invest 59:426 431, 1977 40. PAK CY, PETERSON R: Successful treatment of hyperuricosuric calcium oxalate nephrolithiasis with potassium citrate. Arch Intern Med 146:863 867, 1986 41. GROVER PK, RYAII. RL, POTEZNY N, MARSHALL VR: The effect of

900 Parks et al: Supersaturation and stone composition decreasing the concentration of urinary urate on the crystallization of calcium oxalate in undiluted human urine.j Urol 143:1057 1061, 1990 42. GROVER PK, RYALL RL, MARSHALL VR: Effect of urate on calcium oxalate crystallization in human urine: Evidence for a promotory role of hyperuricosuria in urolithiasis. Clin Sci 79:9 15, 1990 43. GROVER PK, RYALL RL, POTEZNY N, MARSHALL VR: The effect of decreasing the concentration of urinary urate on the crystallization of calcium oxalate in undiluted human urine.j Urol 143:1057 1061, 1990 44. PAK CYC, HOLT K: Nucleation and growth of brushite and calcium oxalate in urine of stone formers. Metabolism 25:665 673, 1976 45. HESS B, KOK DJ: Nucleation, growth, and aggregation of stoneforming crystals, in Kidney Stones, Medical and Surgical Management, edited by Coo L, AVUS MJ, PAK CYC, PARKS JH, PREMINGER GM, Philadelphia, Lippincott-Raven Publishers, 1996, p 3 46. LIESKE JC, Coo L: Urinary inhibitors and renal stone formation, in Kidney Stones: Medical and Surgical Management, edited by Coo L, AVUS MJ, PAK CYC, PARKS JH, PREMINGER GM, Philadelphia, Lippincott-Raven Publishers, 1996, p 65 47. MANDEL OS, MANDEL NS: Crystal-crystal interactions, in Kidney Stones: Medical and Surgical Management, edited by Coo L, AVUS Mi, PAK CYC, PARKS JH, PREMINGER GM, Philadelphia, Lippincott- Raven Publishers, 1996, p 115 48. ScHEIo C, KOUL H, HILL WA, LIESKE JC, TOBACK G, MENON M: Oxalate ion and calcium oxalate crystal interactions with renal epithehal cells, in Kidney Stones: Medical and Surgical Management, edited by COE L, AVUS MJ, PAK CYC, PARKS JH, PREMINGER GM, Philadelphia, Lippincott-Raven Publishers, 1996, p 129 49. Coo L, PARKS JH, STRAUSS AL: Accelerated calcium nephrolithiasis. J.4M.4 244:809 810, 1980 50. MILLMAN S, STRAUSS AL, PARKS JH, COE L: Pathogenesis and clinical course of mixed calcium oxalate and uric acid nephrolithiasis. Kidney mt 22:366 370, 1982 51. LINDBERG i, HARVEY i, PAK CYC: Effect of magnesium citrate and magnesium oxide on the crystallization of calcium salts in urine: Changes produced by food magnesium interaction. J Urol 143:248 251, 1990 52. PAK CYC, OH MS, BAKER S, MORRIS is: Effect of meal on the physiological and physiochemical actions of potassium citrate. J Urol 146:803 805, 1991 53. SAKHAEE K, NICAR M, HILL. K, PAR CY: Contrasting effects of potassium citrate and sodium citrate therapies on urinary chemistries and crystallization of stone-forming salts. Kidney mt 24:348 352, 1983 54. PREMINGER GM, SAKHAEE K, PAK CY: Alkali action on the urinary crystallization of calcium salts: Contrasting responses to sodium citrate and potassium citrate. J Urol 139:240 242, 1988 55. PAR CYC, HOLT K, ZERWEKH JE, BARILLA DE: Effects of orthophosphate therapy on the crystallization of calcium salts in urine. (abstract) Miner Electrol Metabol 1:147 154, 1978 56. PREMINGER GM, SAKHAEE K, SKURLA C, PAR CY: Prevention of recurrent calcium stone formation with potassium citrate therapy in patients with distal renal tubular acidosis. J Urol 134:20 23, 1985 57. BRESLAU NA, PADALINO P, KOK Di, KIM YG, PAK CYC: Physicochemical effects of a new slow-release potassium phosphate preparation (UroPhosK)* in absorptive hypercalciuria. J Bone Miner Res 10:394 400, 1995 58. PAK CY, KOENIG K, KHAN R, HAYNES 5, PADALINO P: Physicochemical action of potassium-magnesium citrate in nephrohithiasis. J Bone Miner Res 7:281 285, 1992 59. NICAR Mi, PETERSON R, PAR CYC: Use of potassium citrate as potassium supplement during thiazide therapy of calcium nephrolithiasis. J Urol 131:430 433, 1984 60. PAR CY, ULLER C, SAKHAEE K, PREMINGER GM, BRITTON : Long-term treatment of calcium nephrohithiasis with potassium citrate. J Urol 134:11 19, 1985 61. SAKHAEE K, HARVEY JA, PADALINO P, WHITSON P, PAR CYC: The potential role of salt abuse on the risk for kidney stone formation. J Urol 150:310 312, 1993 62. BRESLAU NA, BRINKLEY L, HILL KD, PAR CY: Relationship of animal protein-rich diet to kidney stone formation and calcium metabolism. J Clin Endocrinol Metab 66:140 146, 1988 63. TISELIUS H: A simplified estimate of the ion-activity product of calcium phosphate in urine. Eur Urol 10:191 195, 1984