Abstract
Purpose: :
To examine the properties of ocular expansion in myopia with reference to regional variation in ocular dimensions of axial [nasal(N) - temporal(T) meridians] and sagittal [superior(S) - inferior(I) meridians] planes using data derived from in vivo MRI-generated 3D images of the human eye.
Methods: :
Sixty six young adult subjects [35 with myopia MSE (D) range -1.37 to -10.00, mean -5.40±2.69, axial length (AL mm) mean 25.53±1.15; 31 with emmetropia MSE (>-0.50≤+0.75) mean +0.03±0.39, AL mean 23.88±0.82], matched approximately for ethnicity [British South Asian (42%), British White] and gender [female (70%)], were scanned using a Siemens Trio 3-tesla MRI. Vector co-ordinates of 3D surface polygons (N=~32K) were assigned to N, T, S and I quadrants. Quadrant data for right eyes were collapsed about each respective meridian and represented graphically in 2D. Mean data were plotted for each meridian at 5% intervals between 15-75% of axial length (12 data points comprising the anterior region, AR) and 2.5% intervals between 75-100% of axial length (10 data points comprising the posterior region, PR). The coefficient of variation [CoV = (standard deviation/mean)*100] was calculated for each of the 22 data points and transformed to permit 2-way mixed repeated measures ANOVAs. Emmetropic CoV data across AR and PR were used as the ‘control’ condition against which myopic CoV data could be compared for each meridian.
Results: :
Plots of mean data indicated that, generally and for both N:T and S:I meridians, ocular expansion in eyes with myopia was axial such that the oblate conformation that characterised eyes with emmetropia was significantly less oblate. Plots of CoV data indicated clear differences over the AR between myopes (M) and emmetropes (E) for N and T meridians: M vs E [F(1,22)=10.63, p=0.004]; N vs T [F(1,22)=14.90, p=0.001]; interaction [F(1,22)=99.63, p<0.001]. A distinctive feature was that CoV for the myopic T meridian was between 2.12 and 2.98% greater than CoV for the emmetropic T meridian across all AR data points. No significant effects were found for the PR. Although significant differences between S and I were also evident across AR [F(1,22)=16.46, p=0.001] these were independent of refractive status.
Conclusions: :
Variation in ocular dimensions in the temporal meridian of individuals with myopia differs significantly from those with emmetropia and may be a corollary of variation in concomitant structural, biomolecular or neuro-optical responses to myopigenic stimuli.
Keywords: myopia • refractive error development • sclera