Voided volume may not impact stone outcomes: Review of a large institutional nephrolithiasis cohort

Abstract Background Urologic guidelines universally recommend increasing fluid intake for kidney stone prevention. Increased voided volume is thought to help reduce stone recurrence and severity, but supporting evidence is limited. Patients and Methods Nephrolithiasis outcomes and 24‐h urine data for patients from the Registry for Stones of the Kidney and Ureter (ReSKU), a registry of nephrolithiasis patients collected between 2015 and 2020, were retrospectively analysed. Outcome was stone events, either an office visit where a patient reports symptomatic passage of stones or surgery for stone removal. Results We identified 450 stone patients with 24‐h urine and kidney stone outcome data. There was no significant difference in 24‐h voided volume between patients with one stone event and patients with two or more stone events. On multivariable logistic regression, after controlling for age, gender, BMI, and 24‐h sodium and creatinine per kilogram, no significant associations were found between voided volume and stone events. There was a statistically significant negative correlation noted between voided volume and stone events in calcium oxalate dihydrate stone formers (Spearman R = −0.42, p = 0.04), but not others. Conclusions Twenty‐four‐hour voided volume was not associated with stone events in a large institutional cohort, and subset analysis reveals that some stone formers may benefit more from increased voided volume than others; identifying such patients represents a novel precision medicine opportunity.

The most cited study supporting these guidelines is the prospective randomized clinical trial by Borghi et al. in 1996 showing that patients started on a high fluid intake program with a target urine volume of more than 2 L daily had 50% lower stone recurrence rates and longer time to first stone recurrence compared with controls. 4 Although additional studies supporting higher fluid intake for nephrolithiasis prevention have been reported, [5][6][7][8][9] they are limited as pointed out by a recent Cochrane review. 10 The authors revealed that the Borghi study was the only study that met their strict inclusion criteria.
They further determined that the Borghi study was at risk of multiple biases, including selection bias, performance bias, detection bias, and reporting bias.
By identifying patients from a novel UCSF database, the Registry for Stones of the Kidney and Ureter (ReSKU), 11 with outcomes data and 24-h urine data, we sought to more rigorously examine the association between initial 24-h voided volume and reported stone events.

| Study participants
After 2015, patients presenting to the University of California San Francisco (UCSF) Urology clinic for stone disease management were enrolled into the Registry for Stones of the Kidney and Ureter (ReSKU) database, a prospective, high-quality, automated nephrolithiasis registry that extracts data directly from the electronic health record (EHR). The enrolment protocol and other details have been fully described previously. 11

| Outcomes and study variables
Demographic data on patient age, race, gender, and BMI were analysed; 24-h urine data for voided volume, 24-h urine sodium (Na24), a surrogate for dietary sodium intake, and creatinine per kilogram (CrKg24), a surrogate for kidney function, were obtained. The primary nephrolithiasis outcome was number of stone events, defined as a symptomatic passage of stones or a surgery for stone removal. To prevent overcounting due to incomplete clearance of stones due to technical or surgical factors, patients who had a symptomatic stone episode without passage of stones and a stone removal surgery within 90 days of that episode were only counted as a singular stone event.

| Statistical analysis
Nonparametric Mann-Whitney and Kruskal-Wallis analysis of variance tests were used to compare continuous variables between two or three groups, respectively. Chi-squared tests were used to compare categorical variables between two groups. Multivariable logistic regression analysis was performed to determine the impact of voided volume on stone events while controlling for gender, age, BMI, CrKg24, and Na24. Spearman correlations were performed to determine the impact of voided volume on stone events for each primary stone composition. Statistical analyses were performed using R 3.4.0 (R Foundation, Vienna, Austria) and validated with GraphPad Prism v7.

| RESULTS
The number of patients with both outcomes and 24-h urine data in the ReSKU dataset included 450 patients from 2015 to 2020. Demographic and clinical characteristics of the cohort are summarized in There was no significant difference in 24-h voided volume between patients with one stone event and patients with two or more stone events (Kruskal-Wallis p = 0.06; Figure 1). Multivariable logistic regression analysis demonstrated that female gender (OR 1.93, 95% CI 1.25-3.00, p = 0.0031) and higher BMI (OR 1.04, 95% CI 1.00-1.08, p = 0.03), but not voided volume (OR 1.04, 95% CI 0.83-1.33, p = 0.68) were associated with multiple stone events (Table 2). Na24 and Cr24Kg were found to be highly correlated with voided volume (R = 0.31 and 0.22, respectively, p < 0.0001 for both). Additionally, we identified 205 patients within our cohort who have had at least two 24-h urine tests. We found that the average change in voided volume between the two urine collections was minimal ( Figure 2A).
Furthermore, we found no significant difference in the average change in voided volume between patients with one stone event and patients with two or more stone events (Kruskal-Wallis p = 0.82; Figure 2B).
The stone cohort was then divided based on primary stone composition, and correlation analysis was performed to determine whether voided volume may be more predictive of stone events in patients with specific stone compositions. There was a statistically significant negative correlation noted between voided volume and stone events in calcium oxalate dihydrate stone formers (Spearman R = À0.42, p = 0.04; Table 3; Figure 3). Additional significant relationships between voided volume and stone events among other types of stone formers were not observed. A recent Cochrane review investigating the role of increased water intake for the prevention of kidney stones identified only a single randomized clinical trial that met criteria for quantitative analysis. 10 In this randomized controlled study by Borghi et al.,199 participants were randomized after their first calcium stone event to either an increased water intake program, aiming for urine volume of greater than 2 L per day, or not. After a follow-up period of 5 years, the increased water intake group demonstrated 50% lower stone recurrence rates and longer time to first recurrence compared with controls. 4 Although eligible for inclusion in the Cochrane review, the authors noted that the study was limited by small sample size, insufficient information about the randomization and blinding processes, and overall low stone recurrence rate over the follow-up period.
Using a well-described registry of urolithiasis patient outcomes and 24-h urine data, we sought to determine whether voided volume was associated with kidney stone recurrence. No significant differences were found between the 24-h voided volume of patients with single stone events or multiple stone events by initial non-parametric testing, confirmed by multivariable regression modelling which con- We also found in our multivariable regression analyses that BMI is associated with stone recurrence and stone events, which is a welldescribed risk factor in literature. 16 Female gender was also found to be associated with disease severity; although nephrolithiasis is traditionally thought to affect males more than females, recent trends demonstrate increasing proportions of female nephrolithiasis over the past few decades. 17,18 Together, these results suggest that both F I G U R E 3 Correlative analysis between voided volume and stone events, divided by stone type younger patients and female patients in our study cohort may have more severe disease compared with that of the general population.
While this may reflect a shifting demographic trend for nephrolithiasis, this also may reflect the nature of the study centre as a tertiary stone clinic, which attracts a greater proportion of patients with relatively complex stone disease compared with other populations. Finally, we found that voided volume was highly correlated with 24-h urine sodium, which is consistent with prior work demonstrating that increased dietary sodium leads to significantly increased urinary voided volume with no statistically significant change in urinary calcium, oxalate or uric acid. 19 In clinical practice, stone management guidelines recommend that patients limit their sodium intake to improve outcomes, 2 which may inadvertently lead to decrease fluid intake and voided volume. Together, this suggests a nuanced balance between recommending decreased sodium intake and increased voided volumes, which warrants further study.
We would be remiss to not mention the potential impact of the Despite these limitations, the findings provide evidence challenging conventional dogma that increased voided volume leads to decreased stone recurrence and severity. Additional prospective randomized clinical trials and precision medicine studies may further shed light on which patients may benefit from increased voided volume.

| CONCLUSION
In this study, we show that 24-h voided volume is not significantly associated with stone outcomes in 450 nephrolithiasis patients. Subset analysis revealed a moderate negative correlation between voided volume and stone outcomes in calcium oxalate dihydrate stone formers, but not others. These findings suggest that established guideline recommendations of increasing voided volume in all nephrolithiasis patients may be more nuanced than initially thought and that identifying populations that may benefit from increased voided volume represents a logical next step.