Objective: To determine predictors for kidney stones recurrence after ESWL or PCNL.

Material and Method: A retrospective cohort study was conducted at a university hospital. The study cohort was patients aged more than 18 years, diagnosed with kidney stones, who were treated with ESWL or PCNL between 2006 and 2009. Medical files were reviewed for clinical profiles, stone characteristics, composition, type of treatment, presence of stone after treatment, stone reappearance, and related laboratory data. Predictors were determined by a multivariable poisson regression and presented as incidence rate ratios (IRRs) with 95% confidence interval.

Results: From a cohort of 252 patients, 240 who had at least one follow-up and with complete plain kidney ureters and bladder (KUB) film or intravenous pyelogram (IVP) were included in analysis. At three years, the total incidence rate of recurrence was 46 per 1,000 person-months. After a multivariable poisson regression clustering by type of stone composition, independent predictors for stone recurrence were age < 50 years (adjusted IRR = 1.3, 95% CI = 1.2-1.4, p < 0.001), ESWL treatment (adjusted IRR = 2.1, 95% CI = 2.1-2.2, p < 0.001), stones located in lower calyx as compared to renal pelvis (adjusted IRR = 8.7, 95% CI = 2.9-25.9, p = 0.001), multiple stones (adjusted IRR = 5.9, 95% CI = 4.8-7.5, p < 0.001), and stone size larger than 20 mm (adjusted IRR = 1.4, 95% CI = 1.2-1.6, p < 0.001).

Conclusion: After stone removals, patients with these predictors should closely be followed up for regular clinical evaluations.

Keywords: Predictors, Kidney stones, Post treatment, Reformation, Incidence rate, Nephrolithotomy