advertisement

---

Editors Selection IGR 14-2

Risk factors: Caffeine

Comment by Steve Mansberger on:

20998 Caffeine consumption and the risk of primary open-angle glaucoma: A prospective cohort study, Kang JH; Willett WC; Rosner BA et al., Investigative Ophthalmology and Visual Science, 2008; 49: 1924-1931

Find related abstracts

---

Caffeinated beverages cause a short-term increase in intraocular pressure with a decrease in ocular circulation, which may increase the risk of glau-coma. The study by Kang et al. (715) examines the relationship between caf-feine consumption and risk of incident glaucoma. This rigorous cohort study in-cludes 79,120 women from the Nurse's Health Study (NHS) and 42,052 men from the Health Professional Follow-up Study (HPFS). The authors report a follow-up response rate of 95%, a laudable proportion considering almost 30 years of follow-up. For this ancillary study, participants contributed person-time if they were older than 40 years, no history of glaucoma, accurate dietary history, and history of an eye exam within the last two years. The authors controlled for multiple factors such as amount and type of caffeine consumption, ethnicity, body mass index, family history, physical activity, alcohol intake, smoking history, and total fluid intake. They validated the food frequency questionnaire for caffeine intake, and performed a chart review to confirm glaucoma.

When our patients ask us about caffeine intake and glaucoma, we should tell them that caffeine was not associated with glaucoma

The authors found no increased risk of glaucoma with increasing amounts of caffeine. Subgroup analyses found associations between caffeine and those with a family history of glaucoma. Stratifying by smoking status, family history of glaucoma, or history of elevated IOP found statistically significant or borderline results. Consumption of tea seemed to have a protective effect for developing glaucoma. Overall, while these subgroup analyses generate interesting research hypotheses, they should be viewed with caution because multiple subgroup analyses increase the possibility of finding a random, statistically significant result (Type 1 error). Also, the large sample size of the study increases the risk of finding a statistically significant result that is not clinically significant.

The authors point out that this study may not be applicable to other ethnic groups since the study's participants were > 90% Caucasian. Also, the study did not provide information about differences between the ancillary study participants vs. the remainder of the NHS and HPFS. Differences, if found, may further decrease the applicability to the general population. When our patients ask us about caffeine intake and glaucoma, we should tell them that caffeine was not as-sociated with glaucoma. I congratulate the authors for their rigorous and important study.

---

• ← Previous
• Next →

---