Current issues of ACP Journal Club are published in Annals of Internal Medicine


Reduced soft drink consumption led to a reduced risk for urinary stone recurrence in men

ACP J Club. 1993 Jan-Feb;118:15. doi:10.7326/ACPJC-1993-118-1-015

Related Content in the Archives
• Letter: Soft Drink Consumption and Urinary Stone Recurrence in Men

Source Citation

Shuster J, Jenkins A, Logan C, et al. Soft drink consumption and urinary stone recurrence: a randomized prevention trial. J Clin Epidemiol. 1992 Aug;45:911-6.



To determine whether reduced soft drink consumption reduces recurrent urinary stones.


Randomized controlled trial with 3-year follow-up.


72 urology practices.


1010 men (age range 35 to 53 y) who had had ≥ 1 physician-confirmed urinary stone episode and consumed > 160 mL of soft drinks per day. Follow-up was 93%.


Patients were stratified by number of previous stones and soda consumption and randomized. Men in the experimental group (n = 504) were asked to refrain from consuming soft drinks and were given literature describing the association between soft drink consumption and stone disease. Men in the experimental group who reported consuming > 50% of baseline intake on follow-up received additional advice. Control patients (n = 505) were given no study information.

Main outcome measures

Patients filled out questionnaires on beverage consumption and stone recurrence every 6 months for 3 years. All reported stone recurrences prompted a blinded telephone call for details and whether the episode was confirmed by a physician.

Main results

Analysis was by intention-to-treat. 64.6% of the study group were recurrence-free at 3 years compared with 58.2% of the control group (P = 0.023 for the 6.4% difference between groups). Although not part of the original design, a subgroup analysis showed a 15% advantage in 3-year recurrence (P = 0.002) when the avoided soft drink was acidified with phosphoric acid and no advantage when the avoided soft drink was acidified with citric acid or a combination of citric and phosphoric acid. Study group patients consumed a mean of 34% of their baseline soda levels compared with a mean of 88% for control group patients. Those in the study group who had complied (< 680 mL soda/wk) had a higher 3-year recurrence-free survival rate than those who had not complied (69.3% vs 58.9%, P = 0.01).


Among men with ≥ 1 bout of urinary stones who consumed at least 1.1 L/wk of soft drinks, reducing soft drink consumption reduced the risk for stone recurrence. This effect may have been restricted to the avoidance of soft drinks acidified solely with phosphoric acid.

Sources of funding: National Institute of Diabetes and Digestive and Kidney Diseases and the Research Assistance Corporation.

For article reprint: Dr. J. Shuster, Research Assistance Corporation, 2026 N.W. 34th Terrace, Gainsville, FL 32605, USA. FAX 352-392-8162.


The methodologic strengths of the study by Shuster and colleagues are the randomized controlled trial design and the large number of patients from different areas of the United States. The weaknesses are related to the lack of details that may permit exploration of alternative explanations for the finding of an effect of the advice regarding soft drinks. Although not stated in the article, the physicians were blinded to allocation. The success of the blinding was dependent on the unblinded patients' silence. The role of co-intervention is uncertain. Patients may have been given dietary advice regarding calcium, oxalate, flesh protein, vitamin D, and fluid consumption. Strauss and colleagues (1) reported that patients who formed stones and remained free of recurrence had a greater increase in urine volume than those who had recurrences. Treatment with thiazides and allopurinol are effective in decreasing the 3-year probability of recurrent stone disease from 55% in control groups to 15% to 25% in treatment groups in randomized controlled trials (2). The proportion of patients who were prescribed thiazides or allopurinol is not provided. No information is given about pre-existing renal stones. Stone episodes could represent new stone formation or passage of existing stones, the probability of the latter being about 35% over 3 years (3).

The improvement in recurrence-free survival is modest at 6.4% over 3 years but is almost entirely attributable to a decreased consumption of soft drinks containing phosphoric acid, where there was a 15% advantage. While awaiting confirmation of this finding, patients should still be advised to increase fluid intake to prevent stone formation but warned to avoid soft drinks containing phosphoric acid as the sole acidifying agent.

David N. Churchill, MD
St. Joseph's HospitalHamilton, Ontario, Canada

Author's Response

Two weakness were cited in the ACP review. First, concern was expressed that even though physicians were blinded to treatment assignment, some patients could have informed their physicians, perhaps leading to differences in concurrent treatment. We contend that this was unlikely, because patients were recruited at the end of their episode, and were not under the care of the participating urologist unless they reached the study end point, recurrence. The second issue dealt with potential stratification factors or covariates that might have been used. With large trials such as this, this is not an issue of the trial validity, but rather precision. Although randomization takes care of a potential bias of unmeasured variables, knowledge of these variables can make a comparison more precise, thereby gaining more power for the trial from the same number of patients. We stratified on 2 important factors: previous soft drink consumption and whether the stone was new or recurrent. Because we did not collect data on previous medications or stone composition, potentially useful covariates were not available to us. However, by concentrating on data that could be obtained reliably from the patient alone, we were able to keep the study very simple and cost effective. This philosophy is known as “Large Simple Trials,” where you opt to use your resources to recruit more subjects to the trial, instead of collecting more data per patient.


1. Strauss AL, Coe FL, Deutsch L, Parks JH. Factors that predict relapse of calcium nephrolithiasis during treatment: a prospective study. Am J Med. 1982;72:17-34.

2. Churchill DN. Medical treatment to prevent recurrent calcium urolithiasis. A guide to criticalappraisal.Miner Electrolyte Metab. 1987;13:294-304.

3. Glowacki LS, Beecroft ML, Cook RJ, Pahl D, Churchill DN. The natural history of asymptomatic urolithiasis. J Urol. 1992;147:319-21.