Does Hofstetter's equation predict the real amplitude of accommodation in children?

The aim was to determine the distribution and associated factors of accommodative amplitude (AA) in six- to 12-year-old children and compare the results with those calculated using Hofstetter's formula.

In a cross-sectional study in 2015, random sampling was done from urban and rural populations of Shahroud, northern Iran. Participating schoolchildren were examined for manifest, cycloplegic and subjective refraction, as well as uncorrected vision and visual acuity. The AA was measured with Donders' push-up method using a ruler. The near point of convergence (NPC) was also measured.

Of the 6,624 selected children, 5,620 participated in the study and after applying the exclusion criteria, the final analyses were done on data from 5,444 schoolchildren. The mean age of the final sample was 9.24 ± 1.71 years (from six to 12 years) and 53.6 per cent (n = 2,919) were boys. Mean measured AA was 14.44 D (95 per cent confidence interval [CI]: 14.33-14.55). In all age groups, the mean measured AA was less than the predicted mean value calculated with the Hofstetter's equation. Mean measured AA was 14.44 D (95 per cent CI: 14.28-14.59) and 14.45 D (95 per cent CI: 14.29-14.6) in boys and girls, respectively (p = 0.926). AA significantly declined with age (coefficient: -0.18, 95 per cent CI: -0.23 to -0.12, p < 0.001). Mean AA in emmetropic, myopic and hyperopic children was 14.31 D, 17.30 D and 14.87 D, respectively. Older age (coefficient = -0.18), living in rural areas (coefficient = -0.48) and NPC (coefficient = 0.47) inversely related with AA and higher AA was associated with a shift of the spherical equivalent refraction toward myopia (coefficient = -0.41).

The differences among groups with different types of refractive error and high AA in children with myopia are important findings of this study. The results of the present study suggest that Hofstetter's formula provides inaccurate AA estimates in children and thus, the interpretation of this index requires further population-based studies in different racial and ethnic groups.