Original Research

Comparison of urinary risk factors for calcium kidney stones in South African black and white subjects is a strategic approach for gaining insights into local and global stone pathogenesis

Allen L. Rodgers

Received: 12 Jan. 2026; Accepted: 07 May 2026; Published: 17 July 2026

Copyright: © 2026. The Authors. Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: In South Africa, kidney stones are relatively rare in the black population (B) compared to their white compatriots (W). Inter-race studies of urinary stone-risk factors have not identified unequivocal reasons to explain this anomaly.

Aim: The present hypotheses are that differences in urine composition between B and W may help explain this phenomenon and may provide insights into this disease more generally.

Setting: Data analysis of research conducted at the University of Cape Town.

Methods: This is a retrospective study involving urinary data collected between 1999 and 2012. These were analysed for conventional urinary risk factors that were categorised as metabolic, environmental and physicochemical. Statistical comparisons were performed (inter-race, inter-gender and control vs stone former). A relative analogous model was adopted in which B and W were regarded as low- and high-stone-risk groups, respectively.

Results: One thousand and thirty-three cases were investigated. Inter-race differences between controls indicated lower stone risk in B. Lower supersaturation of the calcium phosphate salt brushite (Bru) in B was noted as being important because it was the only physicochemical risk factor showing a difference and because of its mooted role as an initiator of stone formation. Higher supersaturation of Bru in males was also noted. There were no statistically significant differences between B and W stone formers.

Conclusion: Some conventional risk factors for kidney stones have a weaker impact on stone formation in B than in W, possibly because of a different urinary milieu in each group.

Contribution: Urinary Bru might be a key factor in explaining the South African stone anomaly and for gaining insights into stone formation globally.

Keywords: urine composition; brushite; race; renal calculi; stone rarity.

Background

The pathogenesis of kidney stone disease has been investigated for over 70 years.^1,2 The principal focus has been the analysis of urine in stone formers (SF) and healthy controls to identify key urinary risk factors.^3,4 Stone researchers have found differences between these groups with respect to promotory metabolic risk factors (excretion of calcium, oxalate, phosphate) and protective factors (excretion of citrate, magnesium, urate and 24 h volume).^5,6,7,8,9 Urinary pH falls into both categories depending on its value.¹⁰ Upward and downward deviations from the normal pH range pose significant risks for different types of stone. Physicochemical ratios of two or more of these urinary components have also been used as predictors of risk.¹¹ Among these is the ion activity product index (AP index) developed by Tiselius.¹² However, the most widely used physicochemical assessment of stone risk is the urinary supersaturation (SS) of stone-forming salts^13,14 and it is closely related equivalent, the urinary supersaturation index (SI).¹⁵ Despite the utilisation of this array of metabolic and physicochemical risk parameters, results have not been consistent in differentiating between SF and healthy controls.¹⁶

The anomaly of stone disease in South Africa is well known – it occurs in the white group (W) but prevalence in the black group (B) is relatively rare.^17,18,19 Several local studies have identified urinary differences between W and B, but these have involved relatively small numbers of participants, have not differentiated between males and females and have focused mainly on comparisons between healthy subjects of both races.^17,19 Each involved relatively few urine collections. As a consequence, results have not unequivocally identified critical metabolic or physicochemical risk factors to fully explain the anomaly. Similarly, on the global stage, stone occurrence has been found to be lower in B than in W, but none has identified compelling arguments to account for this racial difference.^2,20,21,22

We believe that the South African stone anomaly provides a unique human model for investigating urinary stone risk factors in both race groups, which might provide insights into stone pathogenesis locally and on a wider scale. The present study was undertaken with these goals in mind.

Research methods and design

Study design

Based on the historical rarity of stones in B, we adopted an analogous human model of stone formation in which B was regarded as the relatively low-risk stone group, and W was regarded as the relatively high-risk stone group.

Urine samples

Twenty-four-hour urinary data were retrieved from the database of the Kidney Stone Research Laboratory (KSRL) at the University of Cape Town (UCT). Samples were obtained from stone patients (SF) and non-stone forming volunteers (controls [C]) of both races and both genders, all of whom had participated in stone research projects conducted in the KSRL over a period of several years.

Participants

Inclusion criteria for the study were the availability of 24 h urine data and confirmation of calcium oxalate (CaOx) as the main stone type (> 50% of the overall stone composition) for the cases and the absence of disease (questionnaire) in the healthy volunteers. Stone composition was determined by X-ray powder diffraction.

Urine analyses

Urine was analysed for conventional urinary metabolic, physicochemical and environmental risk factors using methods described elsewhere.²³ These risk factors included the aforementioned urinary promoters and inhibitors as well as sodium (Na), potassium (K) and chlorine (CI) (required for calculation of SS, SI and ionic strength) and creatinine (required for normalisation of data in the calculation of AP index described below). Briefly, Na, K, calcium (Ca) and magnesium (Mg) were analysed using flame atomic absorption spectrometry. Oxalate was determined using oxalate decarboxylase. Citrate was determined by conversion to oxaloacetate using citrate lyase. Inorganic phosphorus was determined using ammonium molybdate, creatinine using picric acid, uric acid (UA) using uricase and chloride using an ion-selective chloride electrode.

Physicochemical risk factors

Urinary ionic strength depends mainly on the concentration of sodium and chloride ions and was determined by computerised calculations.²⁴ Relatively high values are regarded as a risk factor for stones.^24,25,26 The AP(CaOx) index for each urine sample was calculated.¹² It utilises five urine parameters (Ca, oxalate [OX], Mg and citrate (Cit) excretions and 24 h urinary volume), all normalised against urinary creatinine excretion.¹² Lastly, SS and SI of common stone-forming salts, including CaOx, calcium phosphate in the form of brushite (Bru) and UA, were calculated using the computer programs EQUIL 2²⁷ and joint expert speciation system (JESS),¹⁵ respectively. The potency of these programmes originates from their simultaneous utilisation of multiple urinary components and a database of thermodynamic constants to determine chemical speciation (concentrations of free unbound ions) for the ultimate calculation of SS values.^15,27 SS and SI were determined in reiterative computations, usually involving at least 13 urinary factors.²⁸ Because EQUIL 2 and JESS have different operating systems and different databases, absolute output values of SS and SI are not identical, but they serve as extremely useful independent measures of the relative risk of stone formation.

Statistical comparisons

Data were non-Gaussian and were analysed using the Kruskal-Wallis test (non-parametric analyses of variance [ANOVA] Instat 3) and R statistical software in RStudio (Posit team, 2025; R Core Team, 2025). with p < 0.05 being regarded as significant. Relevant comparisons between subgroups for each urinary component were performed. However, there were no data for black female stone formers (BFSF), bearing testimony to the rarity of stones in this group.

Ethical considerations

The database used in this study was constructed from numerous studies, all of which were approved by the University of Cape Town Research Ethics Committee (REC Ref 268/2006; REC Ref 390/2007; REC Ref 250/210; REC Ref 366/2011; HREC Ref 174/2011; HREC Ref 366/2011; HREC Ref 344/2013; HREC Ref 072/2014) and were in accordance with the 1964 Helsinki declaration and its amendments or comparable ethical standards. Informed consent was obtained from all participants.

Results

Subjects

A total of 1113 subjects satisfied the inclusion criteria. These were divided into seven sub-groups: black male controls (BMC, n = 264), black female controls (BFC, n = 16), black male stone formers (BMSF, n = 11), white male controls (WMC, n = 413), white female controls (WFC, n = 212), white male stone formers (WMSF, n = 107) and white female stone formers (WFSF, n = 90). There were no data for BFSF. A list of the statistical comparisons that were performed is shown in Table 1. Mean urinary risk factors are presented in Table 2. These are classified as metabolic, environmental and physicochemical in accordance with classifications advocated by Pak and co-workers.²⁹ Statistically significant differences and p-values for the comparisons listed in Table 1 are given in Table 3. Visual representation of these is shown in Figure 1, Figure 2 and Figure 3.

FIGURE 1: Comparison of metabolic risk factors across all sub-groups. (a) Calcium, (b) oxalate, (c) urate, (d) citrate and (e) pH.

FIGURE 2: Comparison of environmental risk factors across all sub-groups. (a) Volume, (b) sodium, (c) potassium, (d) magnesium, (e) phosphate and (f) chloride.

FIGURE 3: Comparison of physicochemical risk factors across all sub-groups. (a) SS CaOx, (b) SS Brushite, (c) SS uric acid, (d) SI CaOx, (e) SI Brushite and (f) SI uric acid, (g) Ion activity product (AP[CaOx]), (h) Ionic strength.

Inter-race comparisons

Statistically significant inter-race differences occurred almost exclusively between BMC and WMC (Table 3; Figures 1–3). Urinary calcium, UA, potassium, magnesium, phosphate, SS(Bru), SI(Bru) and ionic strength were significantly lower in BMC, while urinary oxalate and AP(CaOx) index were significantly higher in this group. There was only one statistically significant difference between BFC and WFC, namely SI(CaOx), which was significantly lower in the former group.

There were no statistically significant differences observed between black and WMSF in any of the urinary parameters investigated. We did not have data for BFSF, which thereby precluded inter-race comparisons between female SF.

Intergender comparisons within the same race group

There were no intergender differences in any of the parameters measured in black control subjects. Among white control subjects, there were several inter-gender differences (WMC vs WFC): urate, sodium, Mg, phosphate, Cl, creatinine, SS(Bru), SI(Bru), SI(CaOx) and ionic strength were significantly greater in males than in females, while citrate and the AP(CaOx) index were significantly lower. Intergender comparisons between the SF in W did not reveal any differences, while those involving black stone formers could not be computed because of the absence of BFSF in our database.

Controls vs stone-formers comparisons within the same race group

There were only three significant differences between controls and stone-formers for black subjects (BMC vs BMSF: SS(UA) and SI(UA) were lower in controls, while pH was higher. In white subjects (WMC vs WMSF and WFC vs WFSF), there were several differences between the subgroups (10 x differences in females and 4 x differences in males). In females, Mg, oxalate, citrate and urinary volume were greater in controls, while ionic strength, SS(CaOx), SS(Bru), SI(Bru), SS(UA) and SI(UA) were lower. In males, Mg and pH were greater in controls, while SI(UA) and ion activity product index were lower.

Discussion

Applying conventional risk factors in conjunction with our relative analogous model, we propose that lower excretions of metabolic risk factors (calcium and UA), environmental risk factors (potassium, magnesium, phosphate) and lower values of physicochemical risk factors (SS[Bru], SI[Bru] and ionic strength) in BMC compared to WMC are consistent with lower stone risk in the former group. Although these urinary components might independently determine the lower stone risk in BMC, it is their net collective effect (given by the supersaturation indices SS and SI) that is likely to be more crucial, as they depend on the equilibrium concentrations of all species that are present.³⁰ In this regard, the lower SS(Bru) and SI(Bru) values in BMC compared to WMC are particularly noteworthy, as high supersaturation of Bru has been previously mooted as an initiator of calcium stone formation.^31,32 Factors that determine elevated SS(Bru) include relatively higher urinary calcium, relatively lower urinary citrate and relatively higher urinary pH.³³ Although only higher urinary calcium was observed in WMC in the present study, we believe that the evidence described above is sufficiently compelling to implicate the lower supersaturation of Bru in BMC as a contributory factor towards explaining the lower occurrence of stones in black males.

An opposing interpretation arises from our observations of lower Mg excretion, higher oxalate excretion and higher AP (CaOx) index in our BMC group, each of which indicates an increased risk of stone formation in this group according to conventional thinking.^6,12

Thus, in BMC, there are two categories of competing urinary risk factors. This notion is not new per se. Commentators have promulgated the idea that stone formation occurs when there is an imbalance between inhibitory and promotory risk factors.³⁴ In the present case, the first of these (lower supersaturation of Bru in BMC vs WMC) indicates a relatively lower risk for CaOx stone formation in BMC compared to WMC, while the second (lower Mg excretion, higher Ox excretion and higher AP[CaOx] index) indicates a relatively higher risk. Predicting which way the balance favours is difficult. We argue that the driving force for precipitation and ultimate stone formation is stronger for SS than that for individual urinary components, because the former is determined by all chemical speciation present, including those associated with the individual components themselves.

It is of further relevance to note that the lower SS of Bru was confirmed by two independent calculations (SS and SI), lending further confidence to this argument. The higher AP(CaOx) index in BMC need not be considered in this context as its formula includes both Mg and Ox excretion, which has already been discussed. As such, we speculate that the lower SS(Bru) in BMC might override the competing effects of the second group of urinary risk factors and that it might contribute towards the lower occurrence of stones in black males. If this turns out to be true, it will have implications for CaOx stone formation globally as it supports the previously held view that SS(Bru) may be a crucial risk factor in this process described.^31,32

The preceding paragraphs have focused on differences in the urinary concentrations of risk factors between B and W male controls. However, the absence of significant differences in other critical conventional risk factors also warrants discussion. Examples are urinary citrate excretion and urinary pH among the metabolic factors and CaOx supersaturation values (SS and SI) among the physicochemical factors. All are crucial for determining CaOx stone formation and protection depending on their individual excretion levels,³⁵ yet no differences were observed between BMC and WMC. We speculate that perhaps some conventional risk factors are irrelevant in the urine of BMC. This notion is supported by our findings in SF too, as no differences were noted here as well (MMSF vs WMSF).

Further mystery is cast by a previous South African study in which lower excretions of urate and citrate (both of which indicate higher stone risk⁹) were observed in recurrent BMSF compared to WMSF.³⁶ These authors did not explain the anomaly. Unfortunately, there are no other South African studies of this kind, so further comparisons are not possible. However, inter-race studies between SFs on the global stage warrant inspection. Encouragingly, in one such study, authors reported lower SS calcium phosphate (presumably Bru) in BSF than in WSF,³⁷ which again draws attention to our earlier hypothesis that an inter-race difference in SS(Bru) is a key factor for accounting for the relatively lower stone occurrence in B. In another study, lower urinary calcium and UA excretions in B were observed.³⁸ While the former indicates lower risk, the latter indicates increased risk,⁹ as predicted by the inhibitor-promoter imbalance theory described earlier.

The conflict between the South African study, in which several inter-race differences occurred,³⁶ and our own, in which none were observed, is concerning. However, we note that in the former, authors commented that (for various reasons), they suspected that their urinary citrate and calcium values in BSF might be overestimated and underestimated, respectively.³⁶ Interestingly, they also commented that urine values in BSF approached those in WSF in most respects. As such, we tentatively suggest that our findings are credible.

Given the above comments and discussion, we speculate that the impact on stone protection and formation of some conventional risk factors is impaired in black South Africans by the presence of one or more urinary components, which are unidentified at this time. Although evidence in support of this is lacking, the same view is shared by the only other South African study of this nature, in which the possible roles of glycosaminoglycans and macromolecules were invoked.³⁶

Regarding female controls, differences in relative supersaturation (RSS)(Bru) or SI(Bru) between the races (BFC vs WFC) were not observed. However, in a study involving 24-h urine composition in black and white women in the US, authors reported lower SS(Bru) in black women.³⁸ They also found lower supersaturation of CaOx, lower calcium excretion and higher urinary pH in this group and attributed the lower SS(Bru) to the latter two factors. We speculate that the previously reported lower incidence of stones in non-South African black females^38,39 might be attributed to a combination of these factors, albeit that we were not able to test this proposal in the present study, as explained earlier.

Based on the evidence described in the preceding paragraph regarding relative stone risk between female controls in the two race groups, we cannot conclude with the same level of confidence as we did with our male controls that lower SS(Bru) is a crucial factor contributing towards the relatively lower stone occurrence in black females, but this remains a possibility for future consideration.

Regarding inter-gender comparisons, no differences were observed between black controls (BMC vs BFC), but in white controls, several differences occurred (WMC vs WFC), most of which indicated a higher risk in males. The latter finding is consistent with those in several global studies,^40,41,42 as well as in the only South African studies of this kind.^19,43,44 Of these six studies, only one included race as a variable.¹⁹ As explained earlier, we were unable to compare black stone formers because of the paucity of females in this group. Intergender comparisons in white stone formers did not reveal any differences. As there have not been any inter-gender studies comparing stone occurrence in black South Africans, probably because of stone rarity in the latter group, the present study is unique for its findings. As stone occurrence is historically less frequent in South African B, we tentatively suggest (as we did when speculating on the reasons for the absence of inter-race differences between BMC and WMC) that the action of conventional risk factors differs in the two race groups.

The notion of an involvement of Bru in stone initiation arises again in our inter-gender comparisons, in which we found higher SS(Bru) and SI(Bru) in white males (WMC vs WFC). As was the case when we interpreted our inter-race finding of lower SS(Bru) and SI(Bru) values in BMC compared to WMC, it can be hypothesised that the reason for stones being more prevalent in males than in females might be because of higher Bru excretion in the former, making this group more susceptible to stone formation as predicted by our analogous model.

Regarding comparisons between controls and SF, all differences (with one exception) were indicative of lower risk in controls.⁴⁵ The exception – higher Ox excretion in WFC compared to WFSF – is confounding, given that oxalate excretion has been widely regarded for many years as one of the most important risk factors for this type of stone. This outlier is an example of the promoter-inhibitor imbalance theory described earlier.

It is recognised that the absence of data for BFSF severely limits the scope of the inter-race and inter-gender comparisons, which weakens the overarching conclusions about female cohorts. The present study would have benefitted from a larger number of urine samples from this group, but this was difficult to achieve ab initio, given the rarity of stones in the latter.

In conclusion, we believe that the present study is important for several reasons. Firstly, it is the first inter-race and inter-gender investigation of urine composition in SF and non- SF in South Africa and among the first globally. Secondly, it has achieved its principal objective of shedding light on the South African stone anomaly by identifying lower supersaturation of urinary Bru as a likely cause of this phenomenon. In so doing, it has simultaneously achieved its secondary objective of initiating a better understanding of stone formation globally by confirming the previous conjecture that SS(Bru) is a key role player in this process. Thirdly, it has been suggested that the inhibitory-promotory action of conventional urinary risk factors in South African B do not have the same impact as those in white males, possibly because of differences in the urinary milieu of the two groups.

Fourthly, this study shows that the South African stone setting provides a useful human model for future research in this field. Intriguing questions regarding the South African stone anomaly provide opportunities for aspiring researchers in medical and basic science disciplines to contribute towards a deeper understanding of this disease locally and globally. Regrettably, as noted by Lazarus and Kaestner, there is no contemporary epidemiological study of nephrolithiasis in South Africa.⁴⁶ Unless this changes, the aforementioned aspirations are unlikely to be met.

Acknowledgements

Some of the data in this study were originally presented as a poster at the 1st Meeting of the European Association of Urology (EAU) Section of Urolithiasis, held in London on 07–10 September 2011 (Eur Urol Suppl 2011; 10:462–463). The content has since been expanded and revised for journal publication. This republication is done with permission from the conference organisers.

The author wishes to express his gratitude to the South African National Research Foundation, the South African Medical Research Council and the UCT for financial support. Thanks are extended to Dalielah Jappie-Mohamed (Doctor of Philosophy [PhD]) of the Department of Chemistry (University of Cape Town [UCT]) and Innocent Karangwa (PhD) of the Department of Statistical Sciences (UCT) for statistical analyses. The author also wishes to collectively acknowledge the contributions to this body of work made by numerous postgraduate research students, research assistants and technical officers.

The author reported that they received funding from the South African National Research Foundation, the South African Medical Research Council and the UCT, which may be affected by the research reported in the enclosed publication. The author has disclosed those interests fully and has implemented an approved plan for managing any potential conflicts arising from their involvement. The terms of these funding arrangements have been reviewed and approved by the affiliated university in accordance with its policy on objectivity in research.

Allen L. Rodgers: Conceptualisation, Funding, Project administration, Resources, Supervision, Writing original draft, Writing review & editing. The author confirms that this work is entirely their own, has reviewed the article, approved the final version for submission and publication and takes full responsibility for the integrity of its findings.

The author wishes to express his gratitude to the South African National Research Foundation, the South African Medical Research Council and the UCT for financial support.

The datasets used and analysed are available from the author, Allen L. Rodgers, on reasonable request.

The views and opinions expressed in this article are those of the author and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.

References