Purpose: To assess the dose-response relationship of vigorous physical activity (running distance, km·d−1) or cardiorespiratory fitness (meters-per-second pace during a 10-km footrace) to the risk for incident glaucoma.
Design: Prospective epidemiologic cohort study.
Methods: Participant-reported, physician-diagnosed incident glaucoma was compared with distance run per week and 10-km footrace performance in a cohort of 29,854 male runners without diabetes followed prospectively for 7.7 yr. The survival analyses were adjusted for age, hypertension, current and past cigarette use, and intakes of meat, fish, fruit, and alcohol.
Results: Two hundred incident glaucoma cases were reported during follow-up. The risk for reported glaucoma decreased 37% per meter per second increment in a 10-km race performance (P = 0.005). Relative to the least fit men (i.e., slowest, ≤3.5 m·s−1), the risk for incident-reported glaucoma declined 29% in those who ran 3.6-4.0 m·s−1 (P = 0.06), 54% for those who ran 4.1-4.5 m·s−1 (P = 0.001), 51% for those who ran 4.6-5.0 m·s−1 (P = 0.04), and glaucoma was nonexistent among the 781 men who exceeded 5.0 m·s−1 (P = 0.03). The risk for incident, reported glaucoma decreased 5% per kilometer per day run at baseline (P = 0.04), which remained significant when adjusted for the 10-km race performance (5% reduction per kilometer per day, P = 0.04), and both body mass index and race performance (P = 0.04). Baseline hypertension was unrelated to the incident glaucoma.
Conclusions: These data provide preliminary evidence that vigorous physicalactivity may reduce glaucoma risk, which, in the absence of medical record validation, could represent ocular hypertensionin addition to frank glaucoma. Additional follow-up with validation is needed to identify the type of glaucoma affected.
Life Sciences Division, Lawrence Berkeley Laboratory, Donner Laboratory, Berkeley, CA
Address for correspondence: Paul T. Williams, Ph.D., Life Sciences Division, Lawrence Berkeley Laboratory, Donner Laboratory, Berkeley, CA 94720; E-mail: firstname.lastname@example.org.
Submitted for publication November 2008.
Accepted for publication January 2009.