Comparison of Thiazide Versus Alkali for Kidney Stone Recurrence Among Individuals With Hypocitraturia and High Urine pH.

Rationale & Objective: To compare symptomatic stone recurrence among individuals with low urinary citrate and high urine pH prescribed thiazides versus alkali. Study Design: Retrospective cohort study. Setting & Participants: Among Medicare beneficiaries with a 24-hour urine collection obtained between 2010 and 2019, individuals with hypocitraturia (≤300 mg/d for men and ≤350 mg/d for women) and high urine pH (>6.3) did not previously receive thiazides or alkali. Exposures: Receipt of thiazide or alkali therapy within 6 months after the urine collection. Outcomes: The incidence of a subsequent symptomatic stone event (ie, an emergency department visit, hospitalization, or surgery). Analytical Approach: Multivariable Cox modeling, adjusting for demographic and baseline 24-hour urine output parameters. Results: Among 1,270 individuals with a mean age of 62.8 years, 68% women, and a median 3.0-year follow-up, we identified 107 and 1,163 individuals receiving thiazides and alkali, respectively. Comparing those who received thiazides versus alkali, there were respective baseline differences in mean 24-hour urinary citrate levels (195 vs 171 mg/d, P = 0.01) and calcium levels (209 vs 151 mg/d, P < 0.001), but not urine pH (6.7 for both groups). By multivariable analysis, compared with alkali, thiazide use was associated with a lower hazard of an symptomatic stone event (HR, 0.64; 95% CI, 0.43-0.97; P = 0.04). A sensitivity analysis matching individuals 1:1 by baseline calcium and citrate levels showed similar results (HR, 0.44; 95% CI, 0.26-0.77; P < 0.01). Limitations: Missing data, miscoding, and residual confounding. Conclusions: Compared with alkali, thiazides are associated with a reduction in symptomatic stone events. These data suggest that a trial is warranted to compare their effectiveness in patients with alkaline urine.