The objective of the study was to update the evaluation and treatment of idiopathic urolithiasis in children in Western society. A secondary goal was to evaluate patients' compliance with high fluid intake. Over 2 years we prospectively studied children referred to us for idiopathic urolithiasis confirmed radiographically, excluding those with secondary disorders. A metabolic urinalysis, which included calcium, citrate, uric acid, oxalate, cystine, and creatinine, was ordered in all patients. Hypercalciuric patients were first treated with a low-sodium (Na)/high-potassium (K) diet and if hypercalciuria persisted, thiazides or potassium citrate was added. Follow-up ultrasound scans were scheduled every 10-12 months. Urine specific gravity (SG) measured during clinic visits was used to assess compliance with high fluid intake. A survey was sent to pediatric urologists and nephrologists to establish a recommended maximal SG value. Thirty healthy school-aged children served as controls. There were 45 children (24 males, 21 females) aged 10.4+/-2.0 years (median 11.0) studied. Stones were retrieved and analyzed in 28 showing calcium composition in all. Urine chemistry analysis was incomplete in 3, and in the others showed hypercalciuria in 33 (78.6%), hypocitraturia in 1 (2.4%), and normal values in 8 (19.0%). Treatment of 33 hypercalciuric patients consisted of diet alone in 13, potassium citrate in 17, thiazides in 2, and potassium citrate and thiazide in 1. All 33 achieved normocalciuria, apart from 2 who remained mildly hypercalciuric on diet alone. The 12 normocalciuric children were treated by diet modification alone. Follow-up ultrasonography showed no new stones in 36 of 39 patients. In 3, new stone formation was associated with recurrence of hypercalciuria after the potassium citrate dose was lowered or discontinued. Upon their first clinic visit, the urine SG of stone formers (1.021+/-0.007) was significantly higher than the maximum SG recommended by 18 physicians of 1.010+/-0.003 ( P<0.001), and not different from the SG in the control group (1.018+/-0.007). Urine SG at follow-up visits was unchanged in stone formers. We therefore propose a step-wise approach in evaluating children with idiopathic urolithiasis in Western society, in which first only urine calcium is studied. Only if urine calcium is normal, should other chemistries be studied. In many hypercalciuric children, low-Na/high-K diet alone is effective, while in most others the addition of potassium citrate is well tolerated, normalizes calciuria, and protects against new stone formation. Children rarely comply with the recommendation of high fluid intake.