A cohort study was performed of children started on the ketogenic diet for intractable epilepsy from 2000 to 2005 (n = 195). Children who developed kidney stones were compared with those without in terms of demographics, urine laboratory markers, and intervention with urine alkalinization (potassium citrate). Thirteen children (6.7%) developed kidney stones. The use of oral potassium citrate significantly decreased the prevalence of stones (3.2% vs 10.0%, P = .049) and increased the mean time on the ketogenic diet before a stone was first noted (260 vs 149 patient-months, P = .29). The prevalence of kidney stones did not correlate with younger age or use of carbonic anhydrate inhibitors (eg, topiramate or zonisamide) but trended toward higher correlation with the presence of hypercalciuria (92% vs 71%, P = .08). No child stopped the diet due to stones; in fact, the total diet duration was longer (median 26 vs 12 months, P < .001). Kidney stones continue to occur in approximately 1 in 20 children on the ketogenic diet, and no statistically significant risk factors were identified in this cohort. As oral potassium citrate was preventative, prospective studies using this medication empirically are warranted.