Soft drink consumption and urinary stone recurrence: A randomized prevention trial

• Published in: Journal of clinical epidemiology Vol. 45; no. 8; pp. 911 - 916
• Main Authors: Shuster, J., Jenkins, A., Logan, C., Barnett, T., Riehle, R., Zackson, D., Wolfe, H., Dale, R., Daley, M., Malik, I., Schnarch, S.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.08.1992
• Subjects: Actuarial Analysis; Adolescent; Adult; Aged; Beverages - adverse effects; Citrate; Citrates - adverse effects; Citric Acid; Humans; Male; Middle Aged; Patient Compliance; Phosphoric acid; Phosphoric Acids - adverse effects; Recurrence; Soft drinks; Urinary Calculi - etiology; Urinary stone disease; Urolithiasis; Citrate; Phosphoric acid; Urinary stone disease; Urolithiasis; Citric acid; Soft drinks
• DOI: 10.1016/0895-4356(92)90074-W
• ISSN: 0895-4356

The object of this study was to determine if a strong association between soft drink (soda) consumption and recurrence of urinary stone disease, found in an earlier case-control study of adult males, had a causal component. The study sample consisted of 1009 male subjects, who completed an episode of urinary stone disease, who were aged 18–75 at that time, and who reported consuming at least 160 ml per day of soft drinks. Half of the subjects were randomized to refrain from consuming soft drinks, while the remaining subjects served as controls. The intervention group had an observed 6.4% advantage in actuarial 3 yr freedom from recurrence ( p = 0.023 one-sided) over the control group. One important secondary finding was that for those who reported at the time of the index stone that their most consumed drink was acidified by phosphoric acid but not citric acid, the experimental group had a 15% higher 3 yr recurrence-free rate than the controls, p = 0.002, while for those who reported at the time of the index stone that their most consumed drink was acidified by citric acid with or without phosphoric acid, the experimental group had a similar 3 yr recurrence-free rate to the controls, p = 0.55. This interaction was significant, p = 0.019.