Role of ocular accommodation in emmetropisation among highly farsighted infants

To assess the relationship between accommodative lag (defocus) and accommodative response to determine which drives emmetropisation in highly farsighted 3-month-old infants who successfully reached emmetropia in a 15-month period.

35 highly hyperopic (≥+5.00 D most hyperopic meridian) 3-month-old infants (57% female) were enrolled in a clinical trial (NCT03669146) to determine the effect on emmetropisation (reaching <+3.0 D) of partial refractive correction (full correction reduced by 3.0 D) and visual exercises to stimulate accommodation. Refractive error was obtained by cycloplegic (1% cyclopentolate) retinoscopy and accommodation was assessed with the monocular estimation method (MEM) at near (33 cm) and with the PlusOptix PowerRefractor (Plusoptix, Nuremberg, Germany) at distance (6 m) and near (33 cm). The effect of accommodative response and lag on emmetropisation was analysed using a repeated measures regression model of change in spherical equivalent refractive error as a function of accommodative lag and accommodative response.

Greater loss of hyperopia was associated with more robust accommodative response at both distance (measured with the PowerRefractor) and near (measured with MEM). These relationships increased in strength with increasing age, reaching B=-0.32 and B=-0.47 diopters of loss of hyperopia per diopter of accommodative response at 18 months (interaction between response and age p=0.004 and p<0.001 for distance and near, respectively). Linear regression analyses showed no significant associations between change in refractive error and defocus at distance or near.

Contrary to defocus-based models of emmetropisation, greater accommodative response and not hyperopic defocus had the stronger influence on the rate of emmetropisation in hyperopic infants.