Got Stones? Get Milk

Abstract & Commentary

Synopsis: Dietary calcium, phytate, and fluid intake are associated with a decreased risk of symptomatic nephrolithiasis in younger women.

Source: Curhan GC, et al. Arch Intern Med. 2004;164:885-891.

The Nurses’ Health Study II enrolled 116,671 female registered nurses in 1989 and continues to follow them. After completing an initial questionnaire, each participant receives follow-up mailings every 2 years. Curhan and colleagues began asking questions about diet in 1991. This prospective, longitudinal cohort study has already yielded intriguing information about the protective effect of vitamin D on multiple sclerosis1 and the relationship between dietary animal fat and breast cancer.2 Using data from the original Nurses’ Health Study I, this same group of investigators demonstrated that a high intake of dietary calcium in older women appeared to decrease the risk of symptomatic kidney stones. On the other hand, supplemental calcium seemed to increase the risk.3 The current study asked the question, "Is there a relationship between diet and kidney stones in younger women?"

A semiquantitative food inventory assessed average food and beverage intake. Information was obtained about supplemental calcium intake (separately and as part of a multivitamin). After excluding women with a history of kidney stones and those in whom the date of diagnosis could not be confirmed, 96,245 women remained. At the beginning of the study these women ranged in age from 25 to 42. If a subject reported a kidney stone in any of the biennial questionnaires, a follow-up questionnaire was mailed that gathered information about the date of occurrence, symptoms, relevant medical circumstances (for example, inflammatory bowel disease, hyperparathyroidism, hyperthyroidism, or urinary tract infection), and stone composition.

During 685,973 person-years of follow up, 1223 symptomatic kidney stones were reported. Only 5.1% of women reported a chronic illness that could conceivably be related to stone creation. However, 17.5% had a UTI. A family history of kidney stones was present in 36.4%, and gout in 21.1%. Stone composition was known in 439 cases; 87.5% contained calcium, 10.0% were urate. In 95.2%, pain was the presenting symptom.

The subjects were divided into 5 groups based on dietary calcium intake. Dietary calcium intake was directly related to intake of animal protein, sodium, potassium, magnesium, phosphorus, vitamin B6, vitamin C, vitamin D, and fluid. It was inversely related to the intake of sucrose and alcohol. The average consumption of calcium supplements and phytate did not vary across the 5 calcium groups. The age-adjusted relative risk (RR) of kidney stones fell dramatically as dietary calcium increased. The lowest quintile was the reference (RR = 1.00). At the upper quintile the RR = 0.54 (95% confidence interval [CI], 0.45-0.63). The RR rose to 0.73 (95% CI, 0.59-0.90) after adjusting for body mass index and multiple dietary constituents (including supplemental calcium and phytate). Supplemental calcium was not associated with risk of kidney stones.

Other dietary factors were studied, again by dividing the group into quintiles and using the lowest as the reference. In multivariate analysis, 2 were associated with a reduced RR of kidney stones: phytate (RR = 0.63; CI, 0.51-0.78) and fluid (RR = 0.68; CI, 0.56-0.83). Animal protein showed a trend toward protection (RR = 0.84; CI, 0.68-1.04). Sucrose intake was associated stone formation at the highest quintile (RR = 1.31; CI, 1.07-1.60).

Comment by Allan J. Wilke, MD

In the United States, the annual incidence of kidney stones is 7 to 21 per 10,000 population.4 What does this study add to our understanding of this disease (remembering that cohort studies such as this cannot prove causation, only show associations)?

1. That an increased intake of fluid reduces the risk of kidney stones makes sense intuitively and forms the basis of our advice to patients who have just completed their first episode of passing a stone: "I want you to drink water until your urine looks like it came fresh out of the faucet!" (This is such a teachable moment!) A previous study showed that coffee, tea, and wine, but not grapefruit juice,5 reduced risk.

2. Current teaching is that we should instruct our patients to limit their intake of animal protein to reduce urate excretion,6 although a randomized controlled study showed no more benefit to a low animal protein, high fiber, high fluid volume diet than just increased fluid alone.7 The current study showed a trend favoring animal protein as a protective factor. This did not reach statistical significance.

3. Increased dietary calcium, but not calcium supplements, is associated with protection against stone formation. This is somewhat counterintuitive, since most stones contain calcium, but consistent with a previous study in men.8 Why dietary calcium, but not calcium supplements? Curhan et al speculate that dietary calcium binds oxalate in the gut. (Calcium oxalate comprises 70% of kidney stones.) They previously reported that most people take calcium supplements without food or only with breakfast.3 If so, there would be fewer opportunities for calcium to bind oxalate. Another theory is that dairy products contain some other constituent that mitigates the risk.

4. Sucrose intake is associated with a higher risk of kidney stones. At first I thought that is was the result of the subjects substituting soft drinks for milk, but sucrose intake remained a risk in multivariate analysis. Curhan et al note that sucrose promotes urinary calcium excretion.

5. Phytate intake is associated with a reduced risk of stone formation. Phytate (a.k.a. phytic acid, inositol hexaphosphate) is a source of phosphorus from plants. It forms insoluble complexes with calcium in the gut. That would reduce urinary calcium excretion and, presumably, the formation of calcium containing stones. Another theoretical mechanism relies on phytate’s ability to inhibit nucleation of calcium oxalate crystals. Phytate is an antioxidant and is present in high concentration in cereal grains, nuts, legumes, and seeds.

While remembering the caveats that this study’s design can only show associations and that the study population was relatively young females, this study supports dietary recommendations that promote consumption of diary products, fluid in increased volume, cereal grains, nuts, legumes, and seeds, and limits consumption of sucrose.

Dr. Wilke, Assistant Professor of Family Medicine, Medical College of Ohio, Toledo, OH, is Associate Editor of Internal Medicine Alert.


1. Munger KL, et al. Neurology. 2004;62:60-65.

2. Cho E, et al. J Natl Cancer Inst. 2003;95:1079-1085.

3. Curhan GC, et al. Ann Intern Med. 1997;126:497-504.

4. Consensus Conference. Prevention and treatment of kidney stones. JAMA. 1988;260:977-981.

5. Curhan GC, et al. Ann Intern Med. 1998;128:534-540.

6. Parivar F, et al. J Urol. 1996;155:432-440.

7. Hiatt RA, et al. Am J Epidemiol. 1996;144:25-33.

8. Curhan GC, et al. N Engl J Med. 1993;328:833-838.