Andrew Carkeet

Refractive error and monochromatic aberrations in Singaporean children

Higher order optical aberrations were measured in 273 cyclopleged Singaporean school children using a Bausch and Lomb Zywave aberrometer, with 268 of these subjects also undergoing corneal topography measurements (Tomey TMS 2 system). Subjects with low myopia (> -3.00 to -0.50 D) showed slightly, but significantly, less positive levels of spherical aberration than other refractive error groups. Chinese subjects also showed significantly higher amounts of aberrations than Malay subjects, particularly for vertical coma, but also for horizontal coma and spherical aberration. Anterior corneal spherical aberration (calculated from topography) was significantly correlated with whole eye spherical aberration, but did not vary significantly with refractive error or racial background. Residual spherical aberration (i.e. of posterior cornea and crystalline lens) did vary significantly with refractive error and race. Our results do not provide any evidence for aberration-driven form-deprivation as a major mechanism of myopia development.

Modeling logMAR visual acuity scores : Effects of termination rules and alternative forced-choice options

Background. Logarithm of the minimum angle of resolution (logMAR) charts come in a variety of alternative forced-choice (AFC) formats and can be used with a variety of different rules to determine when to stop a subject reading down a chart (termination rules). Methods. Exact calculation and Monte Carlo simulation techniques were used to compare logMAR scores for different termination rules and for infinite-, 26-, 10-, 8-, 4-, and 2-AFC logMAR chart formats. Slope-corrected standard deviation, an index of variability of the underlying ideal threshold, was used as a common metric for the different test conditions. Results. The mean and standard deviation of logMAR scores can be significantly affected by termination rules and AFC format. For different AFC formats, different termination criteria were found to give optimal slope-corrected standard deviations. Conclusion. A number of clinically suitable termination rules are proposed for different AFC format logMAR charts. For letter-by-letter scoring of Bailey-Lovie and EDTRS charts, it is recommended that researchers and practitioners use a termination rule of stopping after four or more mistakes on a line.

Higher order ocular aberrations after cycloplegic and non-cycloplegic pupil dilation.

PURPOSE Clinical aberrometry is commonly undertaken with the use of mydriatic agents, however there is no literature available on whether aberrometry results obtained under cycloplegia differ from those obtained without cycloplegia. METHODS Higher order aberrations were measured over a 6-mm pupil with a Bausch and Lomb Technolas Zywave Aberrometer on the right eyes of 31 young subjects (average age 19.7 +/- 1.7 years; 5 females, 16 males). Two measurement conditions were used for each subject: 1) topical installation of 3 drops 1% cyclopentolate hydrochloride; and 2) topical installation of 1 drop 2.5% phenylephrine hydrochloride, prior to aberrometry measurements. RESULTS For higher order aberrations (3rd to 5th order), average root mean square (RMS) after phenylephrine measurement (0.3852 microm) was significantly lower than after cyclopentolate (0.4259 microm). A small but statistically significant difference was found between the two conditions for average vertical and horizontal coma and, to a lesser extent, horizontal 5th order aberrations. Repeatability RMS, a measure of test-retest measurement repeatability, was similar for the two conditions at 0.15 microm, and significantly lower than the average RMS for the difference between the two conditions (residual RMS) of 0.22 microm. CONCLUSIONS The difference between cycloplegic and non-cycloplegic aberration measurements has implications for surgical correction of higher order aberrations.

Prevalence rates and epidemiological risk factors for astigmatism in Singapore school children.

Purpose. This study examined the prevalence rate of astigmatism and its epidemiological risk factors in Singapore school children. Methods. In a study of school children aged 7 to 9 years old in two schools in Singapore in 1999, a detailed questionnaire was administered to parents regarding reading or close-work habits, past history of close-work, family history, and socioeconomic factors. Cycloplegic refraction was performed five times in each eye. Defining astigmatism as worse than or equal to 0.5, 0.75, and 1 D cylinder in the right eye, the prevalence of astigmatism was calculated. Results. The study population consisted of 1028 children. The prevalence rate of astigmatism (worse than or equal to 1 D cylinder) was 19.2% (95% confidence interval, 16.8 to 21.6). This was not different between genders, ethnic groups, or age (p > 0.05). With-the-rule astigmatism was more common than against-the-rule astigmatism. The prevalence of astigmatism and myopia was 9.8% (95% confidence interval, 8.0 to 11.6). A high AC/A ratio was associated (p = 0.003) with astigmatism, even after exclusion of myopic children. On vectorial analysis, J0 and J45 were associated with the number of hours of playing video games, whereas J45 was also associated with computer use. Only J45 was associated to male gender, a high AC/A ratio, and a family history of myopia. Conclusions. The prevalence rate of astigmatism (≥1 D) was 19%. Playing video games and computer use may be associated with astigmatism severity, although the presence of astigmatism (≥1 D) was not associated with any nearwork factors. A family history of myopia was associated with oblique astigmatism severity. A high AC/A ratio is associated with astigmatism, and this requires further investigation.

Motion Perception in Glaucoma Patients: A Review

Most of the histopathological and psychophysical studies in glaucoma reveal a preferential damage to the magnocellular (M) pathway although a few of them support a damage to the parvocellular (P) pathway as well. In glaucoma, the visual fields are usually evaluated by conventional perimetry. However, it has been demonstrated that 20-40% of ganglion cells are lost before field defects are detected using conventional perimetry. Therefore, new psychophysical tests have recently been designed in order to specifically isolate and evaluate the visual mechanisms that are impaired at the early stages of glaucoma. In this context, several authors have addressed the issue of motion perception under the hypothesis of a predominant damage of the M pathway in glaucoma, and that motion perception is mediated mainly by M pathway. The results of these studies depict a large variation in the percentage of patients showing anomalous motion perception. Overall, motion thresholds are elevated in both glaucoma and ocular hypertensive patients as compared to control subjects, irrespective of the stimulus size and eccentricity. The test which discriminates best between patients and normal subjects is motion perimetry. The visual field defects in glaucoma patients identified by conventional perimetry and motion perimetry are similar, but the sizes of the defects are usually larger with motion perimetry. However, motion tests in central vision have no correlation with visual field defect on conventional perimetry. In glaucoma, loss of performance on motion perception tests does not necessarily support the existence of a specific deficit in the M pathway, because some behavioral studies suggest that the P pathway can also mediate motion perception. It is also difficult to conclude that motion perception is specifically affected in glaucoma because most of these studies do not yield a comparison with other visual functions. Despite these difficulties, localized motion perception tests at eccentricities of more than 15 degrees can be considered as a promising diagnostic tool.

Workstation variables and visual discomfort associated with VDTs

We have investigated the effects of a range of workstation factors upon the visual symptoms experienced by a group of 92 visual display terminal (VDT) users. Subjects in the study kept a diary over five consecutive working days in which they recorded the types of visual and postural symptoms which occurred and the types of work tasks being performed. Each subjects workstation was analysed for screen legibility and stability, discomfort and disability glare, and required head postures. By the use of multiple regression analysis techniques we have considered the relative contribution of these factors to the symptoms reported by the users of these workstations. Screen legibility significantly influenced the occurrence of symptoms of ocular discomfort and vertical head movements significantly affected the incidence of postural/headache symptoms.

Regional variations in binocular summation across the visual field

Binocular summation for a contrast detection task was measured as a function of eccentricity and target size along the horizontal and vertical meridians fur ten young normal subjects. Binocular summation at the fovea was of the order of 1.4 for all target sixes, although there was some inter subject variation. Binocular summation was highest along the vertical meridian. With increasing eccentricity from the fovea, binocular summation for target size I (0.108° projected diameter) decreased, remained relatively constant for target size III (0.431° projected diameter] and increased with increasing eccentricity from the fovea for target size V (1.724 projected diameter). For target sizes I and III, binocular summation was present only when inter ocular differences in sensitivity were under 5 dB, for target size V this relationship did not hold. Influences such as stimulation of corresponding retinal points and cortical representation are considered.

Development of Vernier Acuity in Childhood

Purpose To measure Vernier acuity and resolution development after 3 years of age. Methods Observers were 39 children with normal vision (aged 3 to 12 years), 10 adult observers with normal vision (aged 19 to 24 years), and 7 adults with amblyopia. Vernier acuity and resolution were measured using uncrowded static stimuli and a 3AFC psychophysical paradigm. Curve fitting was used to estimate A2, the age at which thresholds are twice asymptotic levels. Results Vernier acuity was hyperacute (i.e., finer than predicted from foveal cone size or spacing) in 3- to 4-year-old observers, but developed later (A2=5.6 ± 1.5 years) than resolution acuity (A2=2.2 ± 0.9 years). Conclusions Childrens Vernier thresholds are poorer than would be predicted solely from their decreased foveal photon capture. Therefore cortical immaturity may play a role in childrens relative position acuity deficit. R/V ratios (resolution/Vernier thresholds) for the youngest age group are similar to those for adult nonstrabismic amblyopes, but better than for strabismic amblyopes.

Exact parametric confidence intervals for Bland-Altman limits of agreement.

Purpose The previous literature on Bland-Altman analysis only describes approximate methods for calculating confidence intervals for 95% limits of agreement (LoAs). This article describes exact methods for calculating such confidence intervals based on the assumption that differences in measurement pairs are normally distributed. Methods Two basic situations are considered for calculating LoA confidence intervals: the first, where LoAs are considered individually (i.e., using one-sided tolerance factors for a normal distribution); and the second, where LoAs are considered as a pair (i.e., using two-sided tolerance factors for a normal distribution). Equations underlying the calculation of exact confidence limits are briefly outlined. Results To assist in determining confidence intervals for LoAs (considered individually and as a pair), tables of coefficients have been included for degrees of freedom between 1 and 1000. Numerical examples, showing the use of the tables for calculating confidence limits for Bland-Altman LoAs, have been provided. Conclusions Exact confidence intervals for LoAs can differ considerably from the Bland and Altman approximate method, especially for sample sizes that are not large. There are better, more precise methods for calculating confidence intervals for LoAs than the Bland and Altman approximate method, although even an approximate calculation of confidence intervals for LoAs is likely to be better than none at all. Reporting confidence limits for LoAs considered as a pair is appropriate for most situations; however, there may be circumstances where it is appropriate to report confidence limits for LoAs considered individually.

Relationship between goblet cell density and tear function tests

Aim: The objective of this study was to find a relationship between goblet cell density (GCD) and other diagnostic tests of dry eye in a group of normal healthy Chinese subjects. The capability of using GCD as a tear function test was assessed.