Purpose: Currently there are attempts to slow myopia progression by manipulating peripheral refractive error. This study proposed to establish the distribution of peripheral refractive errors in hyperopic, emmetropic and myopic children and to test the hypothesis that relative peripheral hyperopia is a risk factor in the onset and progression of myopia. Methods: Refraction was measured under non-cycloplegic conditions, at 0°, 10° (superior, inferior, temporal and nasal retina) and 30° (temporal and nasal retina), at distance and near. Central spherical equivalent refractive error (SER) was used to classify the eyes as myopic (≤ −0.75 D), emmetropic (−0.75 < SER < +0.75 D) or hyperopic (≥ +0.75 D). Relative peripheral refraction was calculated as the difference between the central (i.e. foveal) and peripheral refractive measurements. At baseline, measurements were taken from 554 children and in a subset of 300 of these same children at the follow-up visit. The time interval between initial and follow-up measurement was 9.71 ± 0.87 months. Results: Results were analysed on 528 participants (10.21 ±0.94 years old) at baseline and 286 longitudinally. At baseline, myopic children (n=61) had relative peripheral hyperopia at all eccentricities at distance and near, except at 10°-superior retina where relative peripheral myopia was observed at near. Hyperopic eyes displayed relative peripheral myopia at all eccentricities, at distance and near. The emmetropes showed a shift from relative peripheral myopia at distance to relative peripheral hyperopia at near at all eccentricities, except at 10°-superior retina, where the relative peripheral myopia was maintained at near. In the longitudinal data analysis, myopes who became more myopic did not show greater relative peripheral hyperopia at baseline compared with myopic sub-groups whose central refraction remained stable. Conclusions: The peripheral refractive profile differences between different refractive groups that are reported in other studies have been confirmed in this study. Relative peripheral hyperopia is not found to be a significant risk factor in the onset or progression of myopia in children.

While prolonged nearwork is considered to be an environmental risk factor associated with myopia development, an underlying genetic susceptibility to nearwork-induced accommodative adaptation may be one possible mechanism for human myopia development. As the control of accommodation by the autonomic system may be one such genetically predetermined system, this research sought to investigate whether an anomaly of the autonomic control of accommodation may be responsible for myopia development and progression. The emphasis of this work was determining the effect of altering the sympathetic input to the ciliary muscle on accommodation responses such as tonic accommodation and nearwork-induced accommodative adaptation in myopes and non-myopes. The first study of the thesis was based on observations of Gilmartin and Winfield (1995) which suggested that a deficit in the sympathetic inputs to the ciliary muscle may be associated with a propensity for myopia development. The effect of ß-antagonism with timolol application on accommodation characteristics was studied in different refractive error groups. Our results support the previous findings that a deficit of sympathetic facility during nearwork was not a feature of late-onset myopia. However it was found that classifying myopes according to stability of their myopia and their ethnic background was important and this allowed differentiation between accommodation responses and characteristics of the ciliary muscle autonomic inputs, with the greatest difference observed between Caucasian stable myopes and Asian progressing myopes. Progressing myopes, particularly those with an Asian background, demonstrated enhanced susceptibility to nearwork-induced accommodative adaptation and this was suggested to result from a possible parasympathetic dominance and a relative sympathetic deficit to the ciliary muscle. In contrast, stable myopes, particularly those with an Asian background, demonstrated minimal accommodation changes following nearwork (counter-adaptation in some cases), and increased accommodative adaptation with ß-antagonism, suggesting sympathetic dominance as the possible autonomic accommodation control profile. As ethnic background was found to be an important factor, a similar study was also conducted in a group of Hong Kong Chinese children to investigate if enhanced susceptibility to nearwork-induced changes in accommodation may explain in part the high prevalence of myopia in Hong Kong. Despite some minor differences in methodology between the two studies, the Hong Kong stable myopic children demonstrated counter-adaptive changes and greater accommodative adaptation with timolol, findings that were consistent with those of the adult Asian stable myopes. Both Asian progressing myopic children and adults also showed greater accommodative adaptation than the stable myopes and similar response profiles following ß-adrenergic antagonism. Thus a combination of genetically predetermined accommodation profiles that confer high susceptibility and extreme environmental pressures is a likely explanation for the increase in myopia over the past decades in Asian countries. The hypothesis that a sympathetic deficit is linked to myopia was also investigated by comparing the effect of â-stimulation with salbutamol, a ß-agonist, on accommodation with that of ß-antagonism using timolol. It was hypothesized that salbutamol would have the opposite effect of timolol, and that it would have a greater effect on subjects who demonstrated greater accommodative adaptation effects, i.e. the progressing myopes, compared to those who showed minimal changes in accommodation following nearwork. Consistent with the hypothesis, the effect of sympathetic stimulation with salbutamol application was only evident in the progressing myopes whom we hypothesized may have a parasympathetic dominance and a relative sympathetic deficit type of autonomic imbalance while it did not further enhance the rapid accommodative regression profile demonstrated by the stable myopes. Characteristics of the convergence system and the interaction between accommodation and convergence were also investigated in the Hong Kong children. No significant differences in response AC/A ratios between the emmetropic, stable and progressing myopic children were found and it was concluded that elevated AC/A ratios were not associated with higher myopic progression rate in this sample of Hong Kong children. However, ß-adrenergic antagonism with timolol application produced a greater effect on accommodative convergence (AC) in stable myopic children who presumably have a more adequate, robust sympathetic input to the ciliary muscle, but had little effect on AC of progressing myopic children. This finding again points to the possibility that the autonomic control of the accommodation and convergence systems may be different between stable and progressing myopia. The primary contribution of this study to the understanding of myopia development is that differences in the autonomic control of the ciliary muscle may be responsible for producing anomalous accommodation responses. This could have significant impact on retinal image quality and thus results in myopia development. This knowledge may be incorporated into computer models of accommodation and myopia development and provides scope for further investigation of the therapeutic benefits of autonomic agents for myopia control.