Physical activity in childhood and the association with myopia in adolescence – The CHAMPS Eye Study

Purpose: Myopia is the most frequent eye disease globally and is usually caused by increasing axial growth of the eye during childhood and adolescence. Lifestyle changes such as reduced physical activity (PA) are thought to be the driving force behind the rapid increase of myopia worldwide. The purpose of this study was to investigate the effect of PA on the development of myopia in a Danish cohort of schoolchildren. The hypothesis was that decreased PA during childhood is associated with increased axial length and myopia. Methods: A prospective study of 198 children from the CHAMPS-study DK cohort (Childhood Health, Activity, and Motor Performance School Study Denmark). PA was assessed with GT3X accelerometer (ActiGraph, Florida, USA) worn for 7 full consecutive days at the period August to October 2010: mean intensity was estimated as counts/min (CPM); and cut off-points for sedentary (SED), light (L), moderate (M), and vigorous (V) PA intensity levels were defined according to Melanson et al. Between March and May 2015 the children were invited and participated in the study conducted at the Department of Ophthalmology, Odense, Denmark. Eye examinations were conducted in 2015 and included autorefraction and keratometry (Tonoref II, Nidek, Japan) in cycloplegia. Axial length (AL) was measured in both eyes using Lenstar LS 900 (Haag Streit, Switzerland). Age- and sex-adjusted linear regression was performed to evaluate the effect of PA on the development of myopia. Results: We had full and valid PA and eye dataset from 198 participants. The mean age at follow-up was 15.5 years (range 14.2-17.5) and 52% were male. Mean axial length were 23.5±0.7mm. The mean cycloplegic spherical refractive error was +0.51±1.48diopter (D), with 17% having myopia defined as spherical error ≤-0.5 D. The mean spherical equivalent (SE) was 0.29±1.46D. Eighteen percent were myopic, defined as SE ≤-0.5 D. In an age- and sex-adjusted linear regression analysis each 10% increment in M-PA-time was predictive of a decrease in AL of 1.2 mm (p<0.01) and an increase in SE of 1.5D (p<0.01). Furthermore each 10% increment in SED-PA prompt a 0.3 mm longer AL (P<0.05) and a -0.4D increment of the SE (P<0.05). Conclusions: We found that an increased level of physical activity was associated with a shorter axial length and refractive error for SED-PA and M-PA, consistent with theory.