The Metabolic Stone Workup & the Evidence Behind Stone Prevention

Executive Summary

All trial data verified against primary sources (PubMed / journal of record), June 2026.

Key Points

• Recurrence prevention rests on a coherent physiologic model (lower the supersaturation of the offending salt) but a surprisingly thin and aging randomized-trial base. The widespread instinct that much of practice is closer to eminence than evidence is largely correct.
• The two largest, most rigorous modern RCTs in common stone disease both returned null primary results: NOSTONE (hydrochlorothiazide, n=416)⁴ and PUSH (behavioral hydration, n=1,658)¹.
• The classic positive trials^2,3,6,7,8,9 are small, mostly single-center, decades old, and several inflate apparent efficacy via historical/pre-treatment control rates — the documented “stone clinic effect”¹³.
• On the added questions: oral calcium with meals lowers urinary oxalate but has no hard-outcome RCT; magnesium supplementation has essentially no supportive RCT (the Mg(OH)₂ arm of Ettinger 1988 was not superior to placebo)⁶; citrate in calcium-phosphate disease is extrapolated and physiologically double-edged; and allopurinol’s randomized evidence is for hyperuricosuric calcium-oxalate stones⁷, not uric-acid stones.
• The only crisp, modern, randomized oxalate data come from the orphan space: lumasiran cut urinary oxalate approximately 65% vs placebo in primary hyperoxaluria type 1 — on a surrogate, not stone events¹⁴.
• By endpoint: one old trial shows fluid reduces symptomatic events²; modern trials capturing radiographic new-stone formation and growth of existing stones were null^1,4; the only positive randomized signal on stone growth is acetohydroxamic acid in struvite¹².

1. The Central Tension: Plausible Physiology, Thin Trials

Kidney-stone prevention is built on an elegant and almost certainly correct physical-chemistry premise: a stone forms when urine is supersaturated with respect to a crystalline salt, and lowering the relative supersaturation of that salt should reduce crystallization. Every standard intervention maps cleanly onto this model — dilute the urine, lower the cation, lower the anion or its precursor, raise an inhibitor, or shift pH out of the salt’s crystallization window.

The difficulty is that physiologic plausibility and supersaturation reduction are surrogate arguments, and the randomized evidence that these maneuvers reduce clinical stone events is far weaker than guidelines imply^10,11. The literature is dominated by small, single-center trials from the 1980s–early 2000s, many comparing on-treatment recurrence to a pre-treatment historical rate — a design biased toward apparent benefit. When the field finally produced large, modern, placebo-controlled trials, the two flagship results were negative^1,4.

2. The Metabolic Workup: Who, What, and When

Stone analysis is the single highest-yield test and should be obtained whenever a stone is retrieved or passed; it reorganizes the differential and frequently overrides the urine chemistry. A basic evaluation (serum calcium, electrolytes/bicarbonate, creatinine, urinalysis with pH; plus stone composition and imaging) suits any stone former. The comprehensive evaluation — two 24-hour urine collections — is reserved by guideline for recurrent, high-risk, or interested first-time formers¹¹, since most single-stone formers will not be shown to benefit from drug therapy.

Two non-consecutive 24-hour collections on the habitual diet are standard, because day-to-day variability is large. Defer collection several weeks after an acute episode or procedure. Serum studies screen for primary hyperparathyroidism, distal renal tubular acidosis (non-anion-gap acidosis with inappropriately alkaline urine), and CKD.

3. Reading the 24-Hour Urine (Litholink-Style Report)

Commercial panels report raw 24-hour excretions, volume and pH, dietary surrogates (sodium, sulfate, urea nitrogen, protein catabolic rate), and — most usefully — calculated relative supersaturations (SS) for calcium oxalate, calcium phosphate (brushite), and uric acid. Read in three passes: (1) check collection adequacy, (2) read the supersaturations to identify the threat salt, then (3) read the analytes to explain why SS is high and what is modifiable.

Collection adequacy first. Confirm completeness using 24-hour creatinine indexed to body weight (approximately 15–20 mg/kg/day men, approximately 10–15 mg/kg/day women). An implausibly low creatinine signals under-collection that spuriously lowers every analyte. Misreading an incomplete collection as “normal calcium” is among the commonest interpretive errors.

Conventional analyte targets (physiologic targets reflected in the AUA framework¹¹, not thresholds validated to predict prevention — treat as directional):

4. Stone Types: Risk Factors and Directed Therapy

4.1 Calcium oxalate (approximately 70–80% of stones)

The workhorse stone and the one nearly all trial evidence actually addresses. Risk factors: low volume, hypercalciuria, hyperoxaluria, hypocitraturia, hyperuricosuria; urine pH is largely irrelevant to CaOx crystallization. Directed therapy: fluid for all; thiazide for hypercalciuria; potassium citrate for hypocitraturia; dietary sodium and animal-protein restriction with maintained normal calcium; oxalate moderation with adequate calcium; allopurinol when hyperuricosuria is present.

4.2 Lowering urinary oxalate: diet, calcium binders, B6, and RNAi

Oxalate can be attacked at intake, gut absorption, and hepatic production — interventions that differ enormously in evidence quality.

• Dietary oxalate restriction + maintained calcium. Lowering high-oxalate foods while keeping normal calcium intake (so gut calcium binds oxalate) is first-line. Borghi 2002 is the supporting signal — oxalate fell on the normal-calcium arm and rose on the low-calcium arm³.
• Oral calcium with meals as an oxalate binder. Calcium supplements taken with meals (calcium citrate is often preferred because it also supplies citrate; calcium carbonate or calcium acetate also bind oxalate) reduce oxalate absorption — the rationale in enteric hyperoxaluria (bariatric surgery, IBD, short gut, fat malabsorption). Evidence is physiologic and observational; small studies show lower urinary oxalate, but there is no randomized hard-outcome (stone) trial. Timing is everything: calcium with meals binds oxalate (helpful); calcium between meals raises calciuria (harmful).
• Bile-acid sequestrants (cholestyramine) are used in enteric hyperoxaluria to bind bile salts/fatty acids and indirectly reduce oxalate absorption — observational only.
• Pyridoxine (vitamin B6) lowers oxalate in a subset of primary hyperoxaluria type 1 (B6-responsive AGXT genotypes); observational, not an idiopathic-stone therapy.
• RNA-interference agents (lumasiran, nedosiran). For primary hyperoxaluria, lumasiran reduced 24-hour urinary oxalate by approximately 65% versus placebo (least-squares mean difference −53.5 percentage points; 84% reaching ≤1.5× ULN vs 0%) in the phase-3 ILLUMINATE-A trial¹⁴; nedosiran is a same-class agent. These are the only methodologically strong, modern, randomized oxalate data — but in a rare genetic disease and on a urinary-oxalate surrogate, not stone events.

4.3 Calcium phosphate (rising, especially in women) — and citrate’s double edge

Driven by higher urine pH (favoring brushite/apatite), hypercalciuria, and hypocitraturia. CaP stones should prompt evaluation for primary hyperparathyroidism, distal RTA, and sometimes medullary sponge kidney. Almost none of the randomized treatment evidence was generated in CaP formers — the trials enrolled CaOx disease — so management is largely extrapolation.

Citrate in CaP disease is the clearest example. The citrate RCTs^8,9 enrolled calcium (predominantly oxalate) stone formers, not a CaP cohort. Mechanistically citrate is double-edged: it corrects hypocitraturia and chelates calcium (good), but the alkali load raises urine pH, which increases calcium-phosphate supersaturation (bad). Net effect in CaP disease is genuinely uncertain, and there is no dedicated randomized trial to settle it.

4.4 Uric acid (approximately 8–10%)

The defining driver is persistently low urine pH (<5.5), usually from impaired renal ammoniagenesis tied to insulin resistance, obesity, and type 2 diabetes; low volume and hyperuricosuria contribute. Uric acid solubility is exquisitely pH-dependent, so the cornerstone is urinary alkalinization to pH approximately 6.5–7.0 (potassium citrate), which can both prevent and chemically dissolve existing uric-acid stones; fluid and metabolic/weight management support this. Allopurinol is reserved for hyperuricosuric or refractory cases (see 6.1).

4.5 Struvite / infection stones (magnesium-ammonium-phosphate)

A microbiologic problem, not a metabolic one: urease-producing organisms (Proteus, Klebsiella, others) hydrolyze urea, alkalinize urine, and precipitate struvite, often as staghorn calculi. Primary therapy is complete surgical clearance (percutaneous nephrolithotomy), because residual fragments harbor organisms and seed regrowth; targeted antibiotics support eradication. The urease inhibitor acetohydroxamic acid (AHA) is the only agent with randomized evidence¹² but is poorly tolerated and adjunctive/palliative.

4.6 Cystine (rare, genetic)

Cystinuria causes high cystine excretion and recurrent stones from childhood. Management: high-volume fluid (often 3–4 L/day), aggressive alkalinization (pH >7–7.5), sodium restriction, and thiol-binding agents (tiopronin, penicillamine) for refractory disease. Evidence is observational; randomized data are minimal given rarity.

5. Does It Actually Work? Evidence for Each Therapy

Each standard intervention paired with its best primary trial, the observed effect, and a candid grade. Pooled relative risks are from the AHRQ/ACP systematic review (28 RCTs)¹⁰.

6. Disaggregating Three Commonly Conflated Therapies

6.1 Allopurinol: two different questions, one answer

Allopurinol is asked to do two unrelated jobs, and only one has randomized support. For calcium-oxalate stones with hyperuricosuria, Ettinger 1986 randomized 60 such patients and found fewer calculi on allopurinol⁷; the AHRQ pooled RR is 0.59, with benefit confined to those with hyperuricemia/hyperuricosuria (mechanism: reducing monosodium-urate–induced salting-out of calcium oxalate). For uric-acid stones, allopurinol lowers urate production and is mechanistically attractive, but there is no randomized trial showing it prevents uric-acid-stone recurrence — alkalinization is the cornerstone, and allopurinol is reserved for hyperuricosuric or alkalinization-refractory disease. So allopurinol’s evidence base is a calcium-oxalate therapy that happens to act on urate, not a uric-acid-stone therapy.

6.2 Magnesium: a popular supplement essentially without a trial

Magnesium complexes oxalate in vitro and is widely sold as a “stone” supplement, but the randomized evidence is negative or absent. In Ettinger 1988, the magnesium-hydroxide arms (650 and 1,300 mg/day) showed reductions from predicted rates that were not significantly better than placebo, while chlorthalidone significantly outperformed both⁶. The frequently cited positive trial⁹ tested potassium-magnesium citrate, and its 85% risk reduction is attributed to citrate (and the combination), not magnesium; there is no quality RCT supporting magnesium monotherapy.

6.3 Citrate in calcium-phosphate disease (recap)

As in 4.3: citrate’s randomized evidence comes from calcium-oxalate cohorts; in calcium-phosphate disease the pH rise can increase CaP supersaturation, there is no dedicated RCT, and use is extrapolated and monitored rather than proven.

7. Preventing New Stones vs Growth of Existing Stones

Stone trials use heterogeneous endpoints, and lumping them overstates the evidence. Separate three outcomes: symptomatic stone events (passage or intervention), formation of new stones on imaging, and growth of pre-existing stones.

Two honest conclusions follow. First, the only positive randomized signal specifically on growth of existing stones is acetohydroxamic acid in struvite¹² — for calcium stones, the trials that rigorously captured radiographic growth (NOSTONE, PUSH) were null^1,4. Second, the older “positive” calcium-stone trials measured new-stone-formation rates against historical/pre-treatment baselines, the design most vulnerable to regression to the mean. The cleaner the endpoint and the more modern the trial, the weaker the apparent benefit becomes.

8. Two Instructive Details the Guidelines Gloss Over

8.1 The famous diet trial did not work the way it is taught

Borghi 2002 is cited as proof that “lowering sodium lowers calcium and prevents stones.” The data are subtler: urinary calcium fell by essentially the same amount (approximately 170 mg/day) in both arms. What separated them was oxalate — it rose in the low-calcium group and fell in the normal-calcium group. The benefit was substantially an oxalate effect mediated by maintained calcium intake, not a clean demonstration that sodium restriction prevents stones by lowering calciuria³.

8.2 The trialists themselves flagged the bias

Two foundational positive trials contain warnings from their own authors. In the allopurinol trial, the placebo group had 63% fewer calculi than its own historical baseline; the authors wrote this “underscores the positive treatment bias that regularly occurs in trials of prophylaxis against renal calculi when historical controls are used”⁷. The chlorthalidone trial made the point more forcefully — every arm, including placebo, showed large reductions from predicted rates, “demonstrat[ing] the need for proper experimental design”⁶. These are confessions embedded in the primary literature.

9. Why the Evidence Is Weaker Than It Looks

• The stone clinic effect / regression to the mean. Patients enroll after a cluster of stones — a statistical peak that regresses regardless of treatment. Hosking 1983 showed 58% of idiopathic calcium formers became metabolically inactive on conservative fluid and diet advice alone, and recommended withholding drugs until this effect is assessed¹³. Trials using pre-treatment or historical control rates attribute this spontaneous regression to the drug.
• Small, old, single-center trials. The positive citrate, allopurinol, and thiazide trials enrolled dozens to low-hundreds at single centers decades ago, with composite (often asymptomatic, imaging-detected) endpoints that inflate effect sizes.
• The modern trials are negative. With large, double-blind, placebo-controlled designs, both flagship results were null — NOSTONE (a companion analysis also found no bone-density benefit)^4,5 and PUSH¹. Citrate and allopurinol have never had a NOSTONE-scale modern replication.
• Surrogate-driven reasoning. Much of practice optimizes 24-hour numbers and supersaturation rather than demonstrated event reduction. NOSTONE is the cautionary tale: thiazide reliably lowers urinary calcium, yet that produced no reduction in recurrence⁴. The RNAi oxalate trials, however rigorous, are also surrogate-endpoint studies¹⁴.

10. PUSH and NOSTONE: The Two First-Line Beliefs That Failed Their Best Trials

Two interventions are recited as the foundation of stone prevention — thiazides for hypercalciuria and water for everyone. Each was finally tested in a large, modern, well-run randomized trial, and each came back null. Because these two beliefs anchor most counseling, they deserve a closer look than a single table row.

10.1 NOSTONE — thiazide, at three doses, did nothing

NOSTONE randomized 416 recurrent calcium stone formers to hydrochlorothiazide 12.5, 25, or 50 mg/day or placebo over a median 2.9 years⁴. The primary composite (symptomatic or radiographic recurrence) was essentially identical across all four arms — no benefit at any dose and no dose-response — even though thiazide did what it is supposed to do and lowered urinary calcium. This is the cleanest cautionary tale in the field: a therapy can reliably move the surrogate (calciuria) and still not move the outcome. A companion analysis found no bone-density benefit either⁵. The older positive thiazide trials that built the guideline recommendation were small, single-center, and decades old.

10.2 PUSH — an intensive program raised urine volume but not stone-free rates

PUSH (Prevention of Urinary Stones with Hydration) was an NIDDK / Urinary Stone Disease Research Network RCT of a behavioral adherence intervention, not of water itself¹. Adults and adolescents (≥12 years) with a symptomatic-stone history and low urine volume were randomized 1:1; both arms received usual care plus a smart water bottle and smartphone app, while the intervention arm added a fluid prescription, financial incentives (approximately $1.50/day for hitting the fluid goal), health coaching, and structured problem-solving. The design and rationale were pre-published¹⁵.

Primary result — negative. At a median 2-year follow-up, symptomatic stone events were 154 (19%) with the intervention vs 165 (20%) with control (HR 0.96, 95% CI 0.77–1.20). There was no difference in stone growth of at least 2 mm, in new radiographic stones, or in the composite of recurrence/new-stone/growth. On safety, no hyponatremia required hospitalization; asymptomatic hyponatremia occurred in 12 (1%) intervention vs 2 (<1%) control participants, and urinary storage symptoms (frequency, urgency, nocturia) were transiently greater in the intervention arm at months 6 and 12.

10.3 The countervailing secondary analysis — and why it is weaker than it looks

A secondary analysis presented at the 2026 AUA meeting (Harper et al., n≈1,461 adults) took a different cut: instead of comparing randomized arms, it compared patients by achieved volume change. Those whose 24-hour volume rose by more than the median (>438 mL/day) had stone events in 29.4% vs 33.3% of the low-change group — adjusted OR 0.76 (95% CI 0.59–0.98, P=.037), roughly 24% lower odds — and the association held even among those who never reached 2.5 L/day. The authors propose urine osmolality as a better target than a fixed volume threshold.

The tension between the two results is the whole lesson. The randomized arm comparison is null (and null even at ≥2.5 L); the observational achieved-volume association is protective. That is a textbook confounding-by-adherence problem: achieved urine volume is a post-randomization variable, so stratifying on it abandons randomization and runs an observational study nested inside the trial. People who successfully raise urine output differ systematically — the "healthy-adherer" trait (the same patients who sustain fluids also take their citrate, keep sodium and oxalate down, and attend follow-up), lower baseline metabolic severity, occupation, and BMI — and several of those differences independently lower recurrence. The canonical warning is the Coronary Drug Project, in which adherent placebo patients had approximately 15% mortality vs approximately 25% in non-adherent placebo patients¹⁶: adherence to a sugar pill "worked." Baseline-covariate adjustment cannot launder out an unmeasured trait or time-varying confounding, so an E-value for OR 0.76 is only on the order of 1.7–2.0 at the point estimate and collapses toward approximately 1.1 near the upper confidence bound — a modest healthy-adherer effect could erase it.

Sourcing note. The PUSH primary results are the peer-reviewed Lancet paper (2026; NCT03244189)¹ and the design paper is Am J Kidney Dis 2021¹⁵. The +438 mL achieved-volume figure and the OR 0.76 achieved-volume association are from the AUA 2026 conference presentation (Harper et al.), not yet a primary peer-reviewed publication — hypothesis-generating, not confirmatory.

11. Practical Synthesis

• Get stone composition and two 24-hour collections in recurrent/high-risk formers; read supersaturations as the dashboard and verify adequacy by creatinine.
• Lead with fluid and diet (normal calcium, lower sodium and animal protein, oxalate moderation, calcium with meals if enteric hyperoxaluria) — low harm, plausible, the only interventions even arguably supported in single-stone formers — and give them a real trial before drugs (the stone clinic effect)¹³.
• Reserve pharmacotherapy for genuinely recurrent/active disease: potassium citrate for hypocitraturia and uric-acid stones, allopurinol only when hyperuricosuria is documented, thiazide for hypercalciuria — while being candid that NOSTONE weakens the thiazide case⁴.
• Do not reach for magnesium (no monotherapy RCT support)⁶ and be cautious with citrate in calcium-phosphate disease (it can raise CaP supersaturation; monitor pH/SS).
• Treat CaP and struvite as their own problems: evaluate CaP for hyperparathyroidism/RTA; manage struvite surgically with organism eradication¹². Refer suspected primary hyperoxaluria for genotyping and RNAi therapy¹⁴.
• Track response with follow-up 24-hour urine / supersaturation and imaging, recognizing you are monitoring a surrogate — and that no calcium-stone therapy has modern randomized proof of halting growth of an existing stone.

References

1. Desai AC, Maalouf NM, Harper JD, et al. Prevention of urinary stones with hydration (PUSH): a randomised clinical trial of an adherence intervention. Lancet. 2026;407(10534):1171–1181. PMID: 41864748
2. Borghi L, Meschi T, Amato F, Briganti A, Novarini A, Giannini A. Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. J Urol. 1996;155(3):839-843.
3. Borghi L, Schianchi T, Meschi T, et al. Comparison of two diets for the prevention of recurrent stones in idiopathic hypercalciuria. N Engl J Med. 2002;346(2):77-84. doi:10.1056/NEJMoa010369
4. Dhayat NA, Bonny O, Roth B, et al. Hydrochlorothiazide and prevention of kidney-stone recurrence (NOSTONE). N Engl J Med. 2023;388(9):781-791. doi:10.1056/NEJMoa2209275
5. Christe A, Primetis E, Cereghetti GM, et al. Hydrochlorothiazide and bone mineral density in patients with kidney stones: post hoc analysis of the NOSTONE trial. Clin J Am Soc Nephrol. 2025;20(5):706-718. doi:10.2215/CJN.0000000659
6. Ettinger B, Citron JT, Livermore B, Dolman LI. Chlorthalidone reduces calcium oxalate calculous recurrence but magnesium hydroxide does not. J Urol. 1988;139(4):679-684. doi:10.1016/S0022-5347(17)42599-7
7. Ettinger B, Tang A, Citron JT, Livermore B, Williams T. Randomized trial of allopurinol in the prevention of calcium oxalate calculi. N Engl J Med. 1986;315(22):1386-1389. doi:10.1056/NEJM198611273152204
8. Barcelo P, Wuhl O, Servitge E, Rousaud A, Pak CYC. Randomized double-blind study of potassium citrate in idiopathic hypocitraturic calcium nephrolithiasis. J Urol. 1993;150(6):1761-1764. doi:10.1016/S0022-5347(17)35888-3
9. Ettinger B, Pak CYC, Citron JT, Thomas C, Adams-Huet B, Vangessel A. Potassium-magnesium citrate is an effective prophylaxis against recurrent calcium oxalate nephrolithiasis. J Urol. 1997;158(6):2069-2073. doi:10.1016/S0022-5347(01)68155-2
10. Fink HA, Wilt TJ, Eidman KE, et al. Medical management to prevent recurrent nephrolithiasis in adults: a systematic review for an American College of Physicians clinical guideline. Ann Intern Med. 2013;158(7):535-543. doi:10.7326/0003-4819-158-7-201304020-00005
11. Pearle MS, Goldfarb DS, Assimos DG, et al. Medical management of kidney stones: AUA guideline. J Urol. 2014;192(2):316-324. doi:10.1016/j.juro.2014.05.006
12. Griffith DP, Gleeson MJ, Lee H, Longuet R, Deman E, Earle N. Randomized, double-blind trial of Lithostat (acetohydroxamic acid) in the palliative treatment of infection-induced urinary calculi. Eur Urol. 1991;20(3):243-247. doi:10.1159/000471707
13. Hosking DH, Erickson SB, Van den Berg CJ, Wilson DM, Smith LH. The stone clinic effect in patients with idiopathic calcium urolithiasis. J Urol. 1983;130(6):1115-1118. doi:10.1016/S0022-5347(17)51711-5
14. Garrelfs SF, Frishberg Y, Hulton SA, et al. Lumasiran, an RNAi therapeutic for primary hyperoxaluria type 1 (ILLUMINATE-A). N Engl J Med. 2021;384(13):1216-1226. doi:10.1056/NEJMoa2021712
15. Scales CD, Desai AC, Harper JD, et al; Urinary Stone Disease Research Network. Prevention of Urinary Stones With Hydration (PUSH): design and rationale of a clinical trial. Am J Kidney Dis. 2021;77(6):898-906.e1. PMID: 33212205
16. Coronary Drug Project Research Group. Influence of adherence to treatment and response of cholesterol on mortality in the Coronary Drug Project. N Engl J Med. 1980;303(18):1038-1041. PMID: 6999345
17. Harper JD, et al. Urine volume change, osmolality, and stone recurrence in the PUSH trial (secondary analysis). Presented at the American Urological Association Annual Meeting, 2026. Conference abstract — not yet a primary peer-reviewed publication.

Citation note: Trial metadata (authors, journal, volume, pages, DOIs, reported results) verified against PubMed and the journals of record. Effect sizes are quoted as reported in the primary articles and the AHRQ/ACP systematic review. Statements of “no RCT” reflect the absence of randomized event-outcome trials for that specific indication as of June 2026.