BACKGROUND: About 10% of people in the United States develop at least 1 symptomatic kidney stone during their lives. The recurrence rate after 10 years is at least 50%. Many physicians recommend a low-calcium diet in patients with calcium oxalate stones to prevent recurrence. Recent studies suggest that a low-calcium diet may not be effective and that intake of animal protein and salt may influence renal calcium excretion. This study compares the traditional low-calcium diet with a diet that is low in animal protein and salt.
POPULATION STUDIED: This study enrolled 120 men with idiopathic hypercalciuria (urinary calcium excretion of more than 300 mg per day on an unrestricted diet) who had been referred to a nephrology clinic in Parma, Italy, and who had had at least 2 episodes of symptomatic renal stones. Reasons for exclusion included previous visits to any “stone disease center” and conditions associated with calcium stones, such as hyperparathyroidism or inflammatory bowel disease.
STUDY DESIGN AND VALIDITY: The investigators randomly assigned subjects, using concealed allocation, to 1 of 2 diets in this randomized controlled study. The low-calcium diet limited calcium intake to about 400 mg per day. The other diet, which included about 1200 mg per day of calcium, limited sodium chloride to about 3000 mg and animal protein to 93 g (15% of total calories). Both groups were advised to limit intake of high-oxalate foods and encouraged to drink 2 liters of water per day in cold weather and 3 liters in warm weather. Subjects were allowed moderate consumption of beer, wine, coffee, and sodas. (Detailed dietary instructions are available to New England Journal of Medicine subscribers in the supplement to the publication at www.nejm.org.) The study followed the patients for 5 years or until they developed clinical or radiologic evidence of a renal stone. Annual x-ray and ultrasound studies identified asymptomatic stone recurrences.
Both groups appeared similar at baseline. Analysis was by intention to treat. The number of withdrawals was similar between groups. Although the subjects could not be masked to their treatment group, the radiologists who confirmed the symptomatic recurrences and diagnosed the asymptomatic recurrences were not aware of the treatment assignments. The authors do not indicate how many of the recurrences were symptomatic.
OUTCOMES MEASURED: The primary outcome was the time to development of the first recurrence of a renal stone, whether or not it was clinically evident. Other outcomes included changes in calcium and oxalate excretion and calcium oxalate saturation in the urine.
RESULTS: After 5 years, the low-protein, low-sodium diet led to fewer recurrences (20% compared with 38% in the low-calcium group, relative risk 0.49, number needed to treat with diet for 5 years = 5.5). The risk of recurrence in the low-calcium group was similar to the 35% to 40% expected in the absence of any intervention. The disease-oriented changes in urine characteristics were predictable: urinary calcium decreased in both groups, but oxalate secretion increased in the low-calcium group, causing greater calcium oxalate saturation.