Tonic accommodation, age, and refractive error in children

PURPOSE. An association between tonic accommodation, the resting accommodat ive position of the eye in the absence of a visually compelling stimulus, a nd refractive error has been reported in adults and children. In general, m yopes have the lowest (or least myopic) levels of tonic accommodation. The purpose in assessing tonic accommodation was to evaluate it as a predictor of onset of myopia. METHODS. Tonic accommodation was measured in children enrolled in the Orind a Longitudinal Study of Myopia using an infrared autorefractor (model R-l; Canon, Lake Success, NY) while children viewed an empty lit field or a dark field with a fixation spot projected in Maxwellian view. Children aged 6 t o 15 years were measured from 1991 through 1994 (n = 714, 766, 771, and 790 during the 4 years, successively). Autorefraction provided refractive erro r and tonic accommodation data, and videophakometry measured crystalline le ns curvatures. RESULTS. Comparison of the two methods for measuring tonic accommodation sh ows a significant effect of age across all years of testing, with the lit e mpty-field test condition yielding higher levels of tonic accommodation com pared with the dark-field test condition in children aged 6 through 11 year s. For data collected in 1994, mean (I-SD) tonic accommodation values for t he lit empty-field condition were significantly lower in myopes, intermedia te in emmetropes, and highest in hyperopes (1.02 +/- 1.18 D, 1.92 +/- 1.59 D, and 2.25 +/- 1.78 D, respectively; Kruskal-Wallis test, P < 0.001; betwe en-group testing shows each group is different from the other two). Age, re fractive error. and Gullstrand lens power were significant terms in a multi ple regression model of tonic accommodation (R-2 = 0.18 for 1994 data). Low er levels of tonic accommodation for children entering the study in the fir st or third grades were not associated with an increased risk of the onset of myopia, whether measured in the lit empty-field test condition (relative risk = 0.90; 95% confidence interval = 0.75, 1.08), or the dark-field test condition (relative risk = 0.83; 95% confidence interval = 0.60, 1.14).