Rock-on or crumble: Do kidney stones pass the test of time? A European Association of Urology Section of Endourology study.

Save

Objective

Kidney stone disease is a relapsing disease and preventive measures are therefore of great importance. A cornerstone for adequate treatment and prevention of kidney stone recurrence is stone analysis. This pilot study aimed to evaluate whether time and different ex vivo environments, such as liquid media or light exposure, influence the morpho-constitutional aspects of kidney stones and whether these changes influence the outcome of X-ray diffraction (XRD) or morphoconstitutional assessment (MCA) for stone analysis.

Methods

Ninety-two stone fragments from 23 kidney stone patients were allocated to four groups (Group A, direct stone analysis; Group B, dry non-transparent environment; Group C, dry transparent environment; Group D, wet non-transparent environment). Fragments from Groups B, C, and D were photographed on postoperative day (POD) 0 and POD 84 and sent for stone analysis afterwards. The objective change in appearance was assessed through the mean Euclidean distance in the Commission Internationale de l'Éclairage L∗a∗b∗ (CIELAB) color space.

The agreement of XRD and MCA between POD 0 and POD 84 was evaluated. Spearman's rank correlation coefficient was calculated to assess the correlation between the objective changes in appearance and the diagnostic accuracy for XRD and MCA.

Results

Overall, the mean Euclidean distance was 5.52 between POD 0 and POD 84. XRD had an agreement between POD 0 and POD 84 of 98%. MCA had an agreement of 92%.

The objective changes in appearance between POD 0 and POD 84 did not correlate with the diagnostic accuracy of XRD or MCA.

Conclusion

The appearance of kidney stones alters over time and in different ex vivo environments but does not change the diagnostic accuracy of XRD or MCA.

Introduction

The global prevalence of kidney stones ranges between 1% and 19% and varies between regions. The lifetime prevalence of kidney stones in Europe, for example, is as high as 14% with a recurrence rate estimated up to 50% after 5 years [[1], [2], [3]]. Furthermore, there is an increase in lifetime prevalence of kidney stones as shown by the National Health and Nutrition Examination Survey data with a prevalence increasing from only 3.2% in 1976–1980 to 8.8% in 2007–2010 [4,5].

Not only do kidney stones have an impact on the health of kidney stone patients, but they also affect their quality of life and represent a significant burden on the economy [6,7]. Data from the United Kingdom showed that the cost of management of urolithiasis is comparable to the initial treatment of bladder and prostate cancer combined [8]. In addition, since kidney stone disease is a relapsing disease with recurrence rates up to 26 per 100 person-years without treatment, adequate metaphylaxis is of great importance [9].

One of the cornerstones for an adequate treatment and prevention of recurrence of kidney stones is stone analysis [10]. Knowing the composition of kidney stones will enable an etiological approach for kidney stone patients. X-ray diffraction (XRD) and infrared spectroscopy (IRS) are the gold standard to accurately identify stone composition and its relative proportions [11].

Morphoconstitutional assessment (MCA), as first described by Daudon et al. [12] in 1993, is an alternative. MCA is based on the morphoconstitutional aspects of kidney stones and their possible underlying pathophysiology [13,14]. More recently, endoscopic stone recognition has been studied by different groups with conflicting results.

XRD, IRS, and MCA are all based on dried, ex vivo kidney stones. In contrast, endoscopic stone recognition is based on wet, in vivo kidney stones [10,15].

Urologists and general practitioners are often confronted with kidney stone patients who have passed calculi prior to their first doctor's visit and have kept these fragments as a souvenir. Therefore, it would be interesting to know if these kidney stones are still useful in the diagnostic process, even after being exposed to ex vivo conditions over time. Although it has been shown that environmental elements (such as heat, moisture, wind, or sunlight) can change the appearance of natural stones, the effect of different ex vivo surroundings on kidney stones is still an underexplored subject [[16], [17], [18]].

This pilot study evaluates whether the time spent under different ex vivo environments, such as fluids or light, influences the morphoconstitutional aspects of kidney stones and whether these changes have an effect on the outcome of XRD or MCA for stone analysis.

Patients and methods

Study design

Between March 2022 and May 2022, kidney stones were collected from 23 patients undergoing endoscopic stone treatment at a tertiary referral center for stone treatment (UMC, the Netherlands). Four stone fragments were retrieved from each patient, resulting in 92 fragments. Per patient, each of those four fragments was then randomly allocated to one of four groups: Group A (fragment sent directly for stone analysis with XRD), Group B (fragment stored in a dry non-transparent container), Group C (fragment stored in a dry transparent container), or Group D (fragment submerged in 0.9% NaCl and stored in a non-transparent container).

After each fragment was anonymized and labeled, fragments from Group A were sent for stone analysis, and fragments from Groups B, C, and D were photographed on postoperative day (POD) 0 and POD 84 under controlled circumstances.

Stones from Group A were sent for analysis with XRD at the Erasmus University Medical Center Rotterdam (Rotterdam, the Netherlands) directly postoperatively to determine their chemical composition. Fragments from Groups B, C, and D were sent for stone analysis with XRD on POD 84.

The photographs were taken with a high-quality camera (Canon EOS 80D, Canon Inc., Tokyo, Japan) mounted on a fixed installation with a stable light source on each side of the camera in a room without natural light to ensure controlled lighting and to avoid shadows (Fig. 1). The fragment was fixed in a Petri dish using silicone to ensure that the position of the stone was identical on POD 0 and POD 84 and thus the same surface of the stone was photographed.

A black dye was added to the silicone in order to create a black background to prevent light reflection. Furthermore, the camera settings and position (ISO setting, shutter time, focus, diaphragm setting, and distance to the stone) were documented and reset for each photograph of the respective fragment to ensure controlled circumstances.Figure 1Setup. (A) The fixed camera and light source; (B) An example of a photograph of a stone fragment.Figure 1

Statistical analysis

SPSS version 28 (IBM Corp., Armonk, NY, USA) was used to perform the descriptive statistical analysis.

The objective changes in appearance of the stone fragments between photographs taken on POD 0 and photographs taken on POD 84 were assessed through a script written in MATLAB version 9.11.0 (R2021b; The MathWorks Inc., Natick, MA, USA) specifically for this study. This script outlined the fragment on the POD 0 photograph and divided this region of interest into equal areas (Fig. 2).

For both time points, the mean color of each area in the region of interest was converted to the Commission Internationale de l'Éclairage L∗a∗b∗ (CIELAB) color space. This is a three-dimensional color space that was defined in 1976 by the Commission Internationale de l'Éclairage. The three dimensions are labeled as L∗, a∗, and b∗.

L∗ stands for lightness where 0 is perfect black and 100 is perfect white, and a∗ and b∗ stand for the color components that correspond to the four primary hues of human vision: more specifically, a∗ is the redness-greenness quality of the color where a positive value represents redness and a negative value represents greenness, and b∗ is the yellowness-blueness quality of the color where a positive value represents yellowness and a negative value represents blueness. The coordinates in the CIELAB color space, or L∗, a∗, and b∗ values, can be calculated from the tristimulus values, where a color under standardized conditions is visually matched against the colors red, green, and blue, and is expressed as X, Y, and Z [19,20]. It is an attempt at providing a perceptually uniform color space in which the distance between two points tells how different the colors are.

Thus, for each area the Euclidean distance between the coordinates on the CIELAB color space of the POD 0 and POD 84 photographs was calculated. The distances of the different areas were averaged, weighted by the number of pixels within the area, to obtain a single number for each stone. The Euclidean distance, the length of the line segment between the two coordinates on the CIELAB color space, expresses the objective color change between POD 0 and POD 84 (Fig.

2).Figure 2Output of the objective assessment of change in color between POD 0 and POD 84. (A) A uric acid stone stored in a dry, non-transparent container; (B) A mixed calcium oxalate–uric acid stone stored in a dry, transparent container. POD, postoperative day; CIELAB, Commission Internationale de l'Éclairage L∗a∗b∗.Figure 2

The XRD agreement was defined as the percentage of cases where the composition of the stone fragments of Group A at POD 0 matched the composition of the stone fragments of Groups B, C, and D at POD 84. Complete agreement was defined as a correspondence in primary, secondary, and, if applicable, tertiary composition, whereas partial agreement was defined as a match of the primary composition of stone fragments between POD 0 and POD 84. Acceptable agreement was defined as a minimum of 80% agreement [21,22].

The MCA agreement between photographs showing the stone fragments on POD 0 and POD 84 was defined as the percentage of cases where an expert in MCA (Daudon M) predicted the same composition based on the morphoconstitutional classification described by Daudon et al. [12,13] and Corrales et al. [14] for both photographs, independent of the actual stone analysis. Photographs were provided for assessment in a random order. The expert was unaware of the results of the stone analysis, the POD, or the environment in which the stone was kept.

Complete agreement was defined as a correspondence in primary, secondary, and, if applicable, tertiary composition, whereas partial agreement was defined as a match of the primary composition of stone fragments between POD 0 and POD 84. The total agreement was the sum of both complete and partial agreements. Acceptable agreement was once again defined as a minimum of 80% agreement [21,22].

Spearman's rank correlation coefficient was calculated to assess the correlation between the objective changes in appearance defined as the Euclidean distance and the diagnostic accuracy for XRD analysis and MCA. A correlation coefficient less than or equal to 0.20 was considered negligible, 0.21–0.40 weak, 0.41–0.60 moderate, 0.61–0.80 strong, and greater than or equal to 0.81 very strong [23].

Ethical approval

This study was performed according to the ethical standards described in the 1964 Declaration of Helsinki and its later amendments. As we used anonymized patient data and this study had no influence on the patients or their treatment, the Medical Research Involving Human Subjects Act (WMO) did not apply to this project. However, we applied for and received a waiver of “no objection” (niet-WMO verklaring) from the Institutional Review Board of the Amsterdam University Medical Centers (W22_101 #22.140.app on April 5, 2022).

Results

Benchmark XRD analysis

The 23 stone fragments from Group A were sent for stone analysis by XRD on POD 0 as a benchmark. This analysis showed seven pure and 16 mixed kidney stones as shown in detail in Table 1.Table 1Results from XRD on postoperative day 0 for stone fragments allocated to Group A (fragments sent directly for stone analysis with XRD).Table 1XRD outcomeValue, nCOM4COM–COD4UA3CHP–CHPD1COM–UA1COM–COD–CHP3COM–COD–CHP–CHPD1COM–CHP1CHP–MAP2CHP–MAP–CMP1COM–CHP–MAP2XRD, X-ray diffraction; CHP, calcium hydroxyphosphate; CHPD, calcium hydrogen phosphate dihydrate; CMP, calcium magnesium phosphate; COD, calcium oxalate dihydrate; COM, calcium oxalate monohydrate; MAP, magnesium ammonium phosphate; UA, uric acid.

Software comparison

The objective changes in appearance of the stone fragments were assessed using the Euclidean distance between coordinates in the CIELAB color space. The Euclidean distance was analyzed for 69 fragments. Overall, a mean distance of 5.52 was found between POD 0 and POD 84.

The highest objective difference was found in Group D, with a mean Euclidean distance of 6.83 (n=23), whereas the stone fragments in Group C had the lowest objective difference with a mean distance of 4.81 (n=23). The stone fragments of Group B had a mean Euclidean distance of 4.99 (n=23). Fig.

2 shows the Euclidean distance for a stone in Group B and one in Group C.

XRD comparison

The results of stone analysis by XRD of Group A fragments were compared with the results of stone analysis by XRD of Groups B, C, and D fragments on POD 84. Of the 69 fragments in Groups B, C, and D, 65 fragments were analyzed. The remaining four fragments were lost in transport (n=2) or did not have enough material for analysis (n=2).

The XRD results of these 65 fragments (sent for stone analysis on POD 84) were compared with the XRD results of their respective fragments in Group A (sent for stone analysis on POD 0).

An agreement (partial or complete) of 98% (n=64) in composition between the fragments of Group A and those of Groups B, C, and D was found by XRD. For fragments of Groups B and C, the agreement was 100% (Group B, n=22; Group C, n=22). Group D had an agreement of 95% (n=20).

This resulted in a high agreement (>80%) for XRD across all three environments. Specifics are presented in Table 2.Table 2Agreement by XRD for stone fragments between those allocated to Group A on POD 0 and those allocated to Groups B, C, and D on POD 84.Table 2AgreementOverall (n=65)Group B (n=22)Group C (n=22)Group D (n=21)Complete agreementa52 (34)55 (12)55 (12)48 (10)Partial agreementb46 (30)45 (10)45 (10)48 (10)No agreement1.5 (1)0 (0)0 (0)4.8 (1)XRD, X-ray diffraction; POD, postoperative day.Note: Group A, fragments sent directly for stone analysis with XRD; Group B, fragments stored in a dry non-transparent container; Group C, fragments stored in a dry transparent container; Group D, fragments submerged in 0.9% NaCl and stored in a non-transparent container; values are presented as % (n).aCorrespondence in primary and secondary (and, if applicable, tertiary) composition.bCorrespondence in primary composition.

Morphoconstitutional comparison

The stone fragments of Groups B, C, and D were photographed on POD 0 and again on POD 84. These photographs were assessed in random order by an expert in MCA. Fifty-one of the 69 photographs were of sufficient quality for MCA.

There was a weak correlation between the Euclidean distance and the quality of the photographs for MCA (correlation coefficient 0.336; p=0.005). An agreement (complete or partial) of 92% (n=47) was found by MCA. More specifically, Group B had an agreement of 89% (n=17), Group C an agreement of 94% (n=17), and Group D an agreement of 93% (n=13).

This resulted in a high agreement (>80%) for MCA across all three environments. Details are shown in Table 3.Table 3Agreement by morphoconstitutional analysis for stone fragments between those allocated to Group A on postoperative day 0 and those allocated to Groups B, C, and D on postoperative day 84.Table 3AgreementOverall (n=51)Group B (n=19)Group C (n=18)Group D (n=14)Complete agreementa80 (41)79 (15)83 (15)79 (11)Partial agreementb12 (6)11 (2)11 (2)14 (2)No agreement7.8 (4)11 (2)5.6 (1)7.1 (1)Note: Group A, fragments sent directly for stone analysis with X-ray diffraction; Group B, fragments stored in a dry non-transparent container; Group C, fragments stored in a dry transparent container; Group D, fragments submerged in 0.9% NaCl and stored in a non-transparent container; values are presented as % (n).aCorrespondence in primary and secondary (and, if applicable, tertiary) composition.bCorrespondence in primary composition.

Spearman's rank correlation coefficient

The objective changes in appearance between POD 0 and POD 84 did not correlate with the diagnostic accuracy of XRD (correlation coefficient 0.041; p=0.7) or MCA (correlation coefficient −0.064; p=0.7).

Discussion

This study evaluated the changes in appearance and composition of kidney stones over time during storage under different ex vivo environments, and whether these different conditions influenced the outcomes of XRD or MCA for stone analysis. We found a high agreement for both XRD and MCA between POD 0 and POD 84. Furthermore, this study found that even though the appearance of kidney stones objectively altered as measured by the Euclidean distance, due to environmental influences, this did not correlate with lower diagnostic accuracy for XRD or MCA.

A recent survey among 471 urologists and residents in urology found that even though urologists find stone analysis important, it is often omitted in the diagnostic process of kidney stone patients [24]. Clinicians are often confronted with patients who had kidney stones in the past and kept the fragments as a souvenir. Therefore, it would be interesting to know whether these kidney stones are useful in the diagnostic process even after being exposed to the test of time and ex vivo conditions [11].

Even though it has been shown that exposure to heat, wind, moisture, or sunlight can change the appearance of natural stones, the effect of different ex vivo environments on kidney stones is still an underexplored subject [[16], [17], [18]]. This could mean that there would be reasons not to use kidney stones for analysis after a certain amount of time, or after exposure to certain ex vivo conditions.

To the best of our knowledge, this is the first study to investigate the effect of different environments on ex vivo kidney stones over time. We observed objective changes in appearance as measured by the Euclidean distance for all three included environments. The largest objective change was found for kidney stones in Group D (kidney stones submerged in 0.9% NaCl, stored in a non-transparent container) with a mean Euclidean distance of 6.83 between POD 0 and POD 84.

Alves et al. [16] performed a review and found that simple contact with water can lead to major disaggregation of natural stones. This effect was also found in our study for kidney stones and is visible in Fig. 3.

This figure shows the effect of 0.9% NaCl on a pure calcium oxalate monohydrate stone (Fig. 3A and B) and on a mixed calcium oxalate–calcium phosphate stone (Fig. 3C and D).

Neither these objective changes in appearance nor the objective changes observed in Groups B and C could be correlated with lower diagnostic accuracy for the chemical composition by XRD or MCA (correlation coefficient 0.041; p=0.7 and correlation coefficient −0.064; p=0.7, respectively).Figure 3Change in appearance between POD 0 and POD 84, with all stones stored in non-transparent containers and submerged in 0.9% NaCl solution. (A) A pure calcium oxalate monohydrate stone on POD 0; (B) A pure calcium oxalate monohydrate stone on POD 84; (C) A mixed calcium oxalate–calcium phosphate stone on POD 0; (D) A mixed calcium oxalate–calcium phosphate stone on POD 84. POD, postoperative day.Figure 3

The major limitation of this study lies within the nature of a pilot study, namely the relatively small number of included kidney stones. Since the most common types of kidney stones can be divided into eight groups (calcium oxalate monohydrate, calcium oxalate dihydrate, calcium hydroxyphosphate, calcium hydrogen phosphate dihydrate, calcium magnesium phosphate, magnesium ammonium phosphate, uric acid, and cystine) and the majority of kidney stones have a mixed composition, the number of possible stone compositions is immense [10,25]. It is therefore not feasible to include all possible compositions in a pilot study.

However, even though the sample size was limited, seven out of the eight main compositions were included in this pilot study, providing an important insight into this topic. Only cystine stones, accounting for 1%–2% of all kidney stones, were not included in this study [11].

Another limitation is the fact that, commonly, only a fraction of each stone is analyzed when XRD is performed, which could lead to sample bias when mixed stones are studied. However, this limitation is inherent to the gold standard for stone analysis described in the European Association of Urology guidelines (XRD and IRS), and is applicable to all studies including mixed stones on this topic [11].

Besides the aforementioned limitations, this study has some evident strengths. The fact that not only the effect of time on stone fragments was assessed, but also different ex vivo environments were assessed, provides important information. Furthermore, the photographs were taken under controlled circumstances leading to a lower chance of bias.

This study used XRD for stone analysis due to availability, and even though it has been shown that the results of XRD and IRS are comparable [26], a study by Taheri et al. [27] found that XRD is more reliable than IRS, strengthening our study design. Finally, another strength of this study is the fact that this study not only evaluated the effect on the gold standard for stone analysis, XRD, but also evaluated the effect on MCA and objective changes such as Euclidean distance in the CIELAB color space.

The aim of this pilot study was to evaluate whether the changes in the appearance of kidney stones due to ex vivo environments and time would alter the results of stone analysis. Based on this pilot study, we are inclined to say that this is not the case.

Conclusion

This pilot study evaluated whether changes in morpho-constitutional aspects of kidney stones due to different ex vivo environments and time influence the outcome of XRD or MCA for stone analysis. Even though the appearance of kidney stones altered through the influences of environmental exposure, this was not correlated with a lower diagnostic accuracy for XRD or MCA. As a consequence, ex vivo preserved kidney stones can be used as a diagnostic tool in the treatment of kidney stone patients.

Author contributions

Study concept and design: Michaël M. E. L.

Henderickx, Simone J. M. Stoots, Guido M.

Kamphuis.

Data acquisition: Michaël M. E. L.

Henderickx, Simone J. M. Stoots, Richard M. van den Elzen, Michel Daudon.

Data analysis: Michaël M. E. L.

Henderickx.

Drafting of the manuscript: Michaël M. E. L.

Henderickx, Simone J. M. Stoots.

Critical revision of the manuscript: Michaël M. E. L.

Henderickx, Simone J. M. Stoots, Richard M. van den Elzen, Michel Daudon, Maurits C.

F. J. de Rotte, Harrie P. Beerlage, Daniel Martijn de Bruin, Wendy P.

J. den Elzen, Guido M. Kamphuis.

Conflicts of interest

The authors declare no conflict of interest.

Related articles

PubMed

The Society of Academic Urology: Match Innovations and Quality Improvement.

Purpose of Review The Urology Match has become increasingly competitive over the past decade, driving rising application numbers, interview inefficiencies, and growing financial and logistical burdens for both applicants and residency programs. The Society of Academic Urology (SAU) provides governance and administrative oversight of the Urology Match and in response, has implemented several quality improvements aimed at increasing transparency, efficiency, and equity across the match process. T...

PubMed

Once again, the highlight of this edition (November-December 2026) of International Brazilian Journal of Urology is robotic surgery.

The November-December number of International Brazilian Journal of Urology presents original contributions with a lot of interesting papers in different fields: Prostate cancer, Robotic Surgery, Kidney cancer, UPJ obstruction, Infertility, Bladder Cancer, Cryptorchidism, Endourology and Reconstructive Urology. The papers came from many different countries such as Brazil, China, Israel, Indonesia, Turkey and Portugal, and as usual the editor ´s comment highlights some of them. The editor in chief...