Data supporting the efficacy of preventive pharmacological therapy (PPT) to reduce urolithiasis recurrence are based on clinical trials with composite outcomes that incorporate imaging findings and have uncertain clinical significance. In a study recently published in AJKD, John Hollingsworth and colleagues evaluated whether the use of PPT leads to fewer symptomatic stone events.

AJKDBlog Interviews Editor Timothy Yau (@Maximal_Change) caught up with Dr. Hollingsworth (@dorkstweet) to discuss real world prevention of urolithiasis with pharmacologic treatment .

John Hollingsworth (@dorkstweet) serves as the Chief of Acute Care Quality and Patient Safety for the Endeavor Health NorthShore Hospitals. He is a Clinical Professor of Surgery at The University of Chicago Pritzker School of Medicine. As an endourologist, his clinical practices focuses on the management of complex urinary stone disease.

AJKDBlog:  Thanks for offering your time for this interview!  It’s very interesting to me to see a study done where you take positive trial data from RCTs and extrapolate whether or not those findings are applicable in the real world, and on patient reported outcomes.  Can you introduce the objective of this study and how it informed your methodology?

Dr. Hollingsworth: Thanks, Tim. As your readership is aware, a recent high-profile multicenter study from Switzerland—the NOSTONE trial—called into question the role of preventive pharmacological therapy, more specifically thiazide diuretics, in kidney stone prevention. There are several plausible reasons for the investigators’ null findings, including 1) statistical power (the Swiss team powered its trial to detect a high reduction rate [50%] in stone events for patients receiving the highest diuretic dose, but the true effect size may be more modest); 2) indication (while contemporary practice guidelines recommend thiazides for patients with underlying hypercalciuria, the Swiss team made no such restriction); and 3) medication nonadherence (nonadherence, which negatively correlates with effectiveness, was high in the NOSTONE intervention group). In this context, we designed our real-world observational study to address some of these trial limitations.

AJKDBlog:  Can you tell us about the Litholink database?  There are several companies that offer stone risk analyses, and even some institutions and hospitals offer their own panel.  Although that isn’t a big focus of this paper, can you tell us some of the data that Litholink offers in their 24 hour urine collection, and which risk factors you chose to identify in this study?

Dr. Hollingsworth: Sure, Litholink is a large central laboratory that, among other things, processes 24-hour urine collection specimens. A subsidiary of LabCorp, Litholink reports on a number of stone risk (24-hour urine volume, calcium, oxalate, citrate, uric acid, and urine pH and calcium oxalate, calcium phosphate, and uric acid supersaturations) and dietary values (24-hour urine sodium, potassium, magnesium, phosphorus, ammonium, chloride, sulfate, and urea nitrogen and protein catabolic rate), as well as normalized urine values. For our study, we used a novel linkage of these 24-hour urine data with medical and pharmacy claims for Medicare beneficiaries who had at least one collection processed by Litholink between 2011 and 2016. This database allowed us to examine healthcare utilization among these beneficiaries, enriched by their urine chemistry results.

AJKDBlog:  So based on the urine chemistries abnormalities mentioned above, you then used Medicare Part D enrollment claims and identified patients that were placed on preventive pharmacologic therapy (PPT) with thiazides (for hypercalciuria), alkali (for hypocitraturia or low urine pH), and uric acid lowering agents (for hyperuricosuria).  How did you define medication adherence, and what did you find when you looked at the data?

Dr. Hollingsworth: So, to measure whether a patient adhered to the medication prescribed, we calculated the proportion of days covered (or PDC). This measure is widely used in the pharmacoepidemiology literature and looks at the proportion of days in which a person has access to a prescribed medication over a given period of interest. When we looked at the data, we found that adherence varied by therapy type and was highest for thiazide diuretics (66%) and lowest for alkali therapy (about 37%). We suspect that this finding relates to patients on alkali are frequently asked to take multiple doses per day.

AJKDBlog: Tell us what the primary outcome you were measuring was, and how you captured symptomatic stone events?

Dr. Hollingsworth: Our primary outcome was the occurrence of a symptomatic kidney stone event. We defined this as an emergency department visit or hospitalization for a diagnosis of kidney stone or kidney stone-directed surgery. Many of the prior trials on preventive pharmacological therapy have used other composite outcomes that were largely driven by imaging findings (e.g., radiographic recurrence and/or stone growth), for which consensus is lacking on the clinical importance. At best, these are surrogates for things about which patients, and the clinicians who manage them, care. This was our motivation for the outcome that we chose.

AJKDBlog: Great, let’s jump into the data now.  What were the big takeaways when it came to what medications were prescribed and the adherence to these regimens.

Dr. Hollingsworth: Out study included nearly 14,000 patients, 31% of whom were prescribed guideline-concordant preventive pharmacological therapy. The remaining were untreated. The most common urine chemistry abnormality was hypocitraturia (54%). Among those prescribed therapy, alkali monotherapy was most common, followed by thiazide diuretic monotherapy. Combination therapy was infrequently prescribed (around 6%). As previously mentioned, adherence rates were highest for thiazide diuretics and lowest for alkali.

AJKDBlog: One of the basic tenets of urolithiasis that trainees are taught is that patients who have formed stones are likely to form them again.  What did you find with regards to recurrent stone events over the 2 year window, and the impact of PPT on them?

Dr. Hollingsworth: Agreed. I tell my patients that urolithiasis is a chronic disease marked by acute exacerbations. And just like other chronic diseases, prevention plays an important management role. To this end, when we performed our time-to-event analysis, we found that patients with hypercalciuria or low urine pH who were prescribed a thiazide diuretic or alkali therapy, respectively, and adhered to their medication, had a significantly lower hazard of experiencing a symptomatic kidney stone event than untreated patients. The benefit of preventive pharmacological therapy in these two patient populations was primarily driven by fewer emergency department visits after initiating therapy. However, no such association was observed for patients with hypocitraturia or hyperuricosuria.

AJKDBlog: In your opinion, what is the impact of these findings on existing clinical guidelines for urolithiasis.  Do they support the evidence for PPT for stone recurrence?

Dr. Hollingsworth: My team fears that enthusiasm for thiazide diuretics has been dampened by the NOSTONE results, which is why data from studies like ours are critical. While observational designs have clear limitations (e.g., omitted variable bias), we took steps to address many of the Swiss trial’s weaknesses and were able to show a protective effect associated with thiazide and alkali use in patients with hypercalciuria and low urine pH. Collectively, our findings provide support for current guideline recommendations, regarding the role of these medications in stone prevention.

AJKDBlog: One of the other core tenets of stone management is that hydration is the first-line intervention and potentially the most important thing patients can do to prevent stone recurrence.  How should the findings from this study be interpreted with this context?

Dr. Hollingsworth: Completely agree. I similarly counsel my patients on the importance of hydration for purposes of stone prevention with a urine output goal of two liters or more daily. I view preventive pharmacological therapy as an adjunct to this. That being said, hydration will result in more frequent voiding, and there are some occupations (e.g., taxi operator) where this could be impractical. For these scenarios, thiazides and alkali therapy in the patient with hypercalciuria and low urine pH, respectively, could be first-line.

AJKDBlog: Any closing thoughts, future stone studies you would like to see, and final takeaway points?

Dr. Hollingsworth: For purposes of monitoring response to therapy, we need studies to help define the target for urine calcium reduction with thiazide diuretics. Moreover, NOSTONE did not look at longer acting thiazide-like diuretics (indapamide, chlorthalidone). Studies elucidating their role would also be important.

AJKDBlog: Thank you for taking the time to do this interview!

To view Hollingsworth et al (subscription required), please visit AJKD.org:

Title: Real-World Effectiveness of Preventive Pharmacological Therapy in Patients With Urolithiasis: A Retrospective Cohort Study
Authors: John M. Hollingsworth, Mary K. Oerline, Ryan S. Hsi, Joseph J. Crivelli, Noah Krampe, John R. Asplin, Vahakn B. Shahinian
DOI: 10.1053/j.ajkd.2023.12.015