A total of 121 normal subjects who visited the hospital for a medical checkup from March to December, 2003 enrolled in this study. The subjects consisted of 68 women and 53 men. The mean age of the subjects was 43.20±13.90 years (Table 1).

The definition of a normal eye was the absence of any ocular or systemic diseases that could affect the RNFLT such as diabetic mellitus or hypertension, a corrected visual acuity over 0.6 by Hahn's visual acuity chart, a spherical equivalent on refraction of within ±3.0 diopters and no anisometropia. Intraocular pressure was measured at 3 times lower than 21 mmHg, the cup-disc ratio was less than 0.6 or the difference of the cup-disc ratio was less than 0.2 between both eyes, and the RNFLT was within 95% of normal range as measured by OCT. Subjects who had a history of ocular trauma or surgery and a family history of glaucoma were excluded.

The RNFLT was measured with a circle scanning area of 3.4mm diameter around the optic disc with a STRATUS Model 3000 OCT (Zeiss-Humphrey, San Leandro, CA., USA). We used a scan profile of fast glaucoma RNFL thickness 3.4 type. After the RNFLT was measured with a circle scanning diameter of 3.4 mm around the optic disc, average values of RNFL thickness were calculated in quadrants and each 30 degree sector. The asymmetry of RNFLT for each eye as well as between the right and left eye was evaluated. To investigate the asymmetry of RNFLT with respect to the horizontal and vertical meridians for individual eyes, the RNFLT was measured for 4 equal 90-degree quadrants (Fig. 1-A) and for 12 equal 30-degree sectors (Fig. 1-B). The differences of RNFLT between the paired quadrants or sectors in individual eye with respect to horizontal and vertical meridians were calculated. To investigate the asymmetry of the RNFLT between the right and left eye, calculations of the differences in the RNFLT of quadrants or sectors between the right and left eye were made.

A correlation between the RNFLT and refractive error was evaluated. The paired t-test was used to determine statistically significant differences in RNFLT of the paired quadrants or sectors in each eye, as well as between the right and left eye. The correlation of the RNFLT with refractive error in each quadrant or sector was assessed by Pearson's correlation coefficients. A p-value less than 0.05 was considered statistically significant.

Although standard visual field perimetry has become the accepted clinical tool for evaluating the effect of disease on visual field threshold, inherent limitations exist in the method. These include subjectivity of the patient's response, learning effect, inability of some patients to maintain central fixation during the thresholding task and higher variability of the threshold determination in more peripheral locations and in damaged locations of the visual field. Many forms of optic neuropathy including glaucoma are difficult to diagnose at an early stage of damage, when optic nerve function and appearance may still be normal by current standards. The problem exists as to how to detect optic neuropathy earlier in its course, at a point when medical or surgical intervention may still save vision.

Recently, the analysis of RNFLT by OCT has become widely used in diagnosing an early glaucomatous change. Although the reliability and reproducibility in measuring and quantifying the RNFLT around the optic nerve head with OCT provide a more objective database concerning the status of glaucomatous change,9,10 there are some difficulties in determining which degree of the RNFLT is included in the normal or abnormal range because a normal distribution of RNFLT in Korean people has not yet been presented. In addition, fairly large normal variations in the morphology of the optic nerve head may be confused with the changes of glaucoma. The normal range of RNFLT according to the patient's age, refractive error and axial length, which can all influence the measurement of RNFLT with OCT, must be considered. It is essential, therefore, to evaluate a patient by an individualized approach. To reveal subtle damage of the optic nerve head, comparison of the RNFLT with respect to the horizontal or vertical meridian for each eye as well as between the right and left eye can be an additional diagnostic tool.

The asymmetry of the optic nerve head and RNFLT has not yet been addressed, although the physiologic cups of the optic nerve head tend to be symmetric between the two eyes of the same individual, with a cup-to-disc ratio difference of greater than 0.2 between fellow eyes occurring in only 1% to 6% of the normal population.11,12

Several studies suggesting the asymmetry of RNFLT using scanning laser polarimetry or histomorphometry have been published.7,8,13-15 Their results were in agreement with the conventional findings of RNFLT known as the "ISN'T rule". In one study that examined RNFLT in normal eyes with the scanning laser polarimetry, the asymmetry of the RNFLT in normal subjects was shown in greater detail, demonstrating that the RNFLT is thicker on the inferior side of the temporal retina and thicker on the superior side of the nasal retina.7,8 In addition, they discovered that the RNFLT was significantly thicker in the right than in the left eye for the temporal quadrant, and thicker in the left eye than in the right eye for an inferior-nasal sector. Although the present study demonstrated that the RNFLT in normal subjects is not symmetrical, there are some inconsistencies of the asymmetric RNFL area with their results. According to our study results, the distribution of RNFLT around the optic nerve head agreed with an "ISNT rule" in the analysis of RNFLT measured with OCT. This study demonstrates that there were significant differences in the superior and inferior sectors with respect to the horizontal meridian. In the RNFLT of sectors that were located at the vertical poles and temporal side to those poles, the RNFLT of inferior sectors was thicker than that of superior sectors. Moreover, it is interesting that these sectors have greater sensitivity to glaucomatous damage. In the RNFLT of sectors located at the nasal side to the vertical poles and close to the horizontal meridian, the RNFLT of the superior sector was thicker than that of the inferior sector (Table 3) (Fig. 2-A). There were significant differences between the nasal and temporal sectors with respect to the vertical meridian. The RNFLT of the nasal sectors close to the horizontal poles was thicker than the temporal sectors. In the RNFLT of sectors located close to the vertical meridian, the RNFLT of the temporal sectors was thicker than the nasal sectors (Table 4) (Fig. 2-B). This study also demonstrates a significant asymmetry of RNFLT between the right and left eye. The RNFLT was thicker in the right than the left eye at the nasal sectors as well as the temporal sectors. As mentioned above, these findings showed some inconsistencies with previous studies.7,8 The reason for the significant difference in the RNFLT between the right and left eye is unresolved. Normal subjects may have various significant asymmetries in the topography of the retinal ganglion cell and corresponding visual function. The topographical asymmetry of retinal ganglion cell density in the human retina was not thoroughly investigated. Further investigations will be necessary to answer this question. Regarding these results, asymmetries of RNFLT should be taken into consideration when evaluating and comparing the RNFLT for each eye as well as between the right and left eye.

In the present study, the parameters that can alter the results of the RNFLT analysis measured with OCT were questioned. These parameters include age, degrees of refractive errors and axial length. It was stated in the previous studies that the RNFLT evaluated by the nerve fiber analyzer was decreased with increasing age in the superior and inferior regions.16 It was reported that almost 5,000 nerve fibers are lost annually as the result of aging,15 although several controversies exist.17,18 However, we could not demonstrate a correlation between the RNFLT and age or axial length due to a limited study group which did not present with a normal distribution of age or axial length. Our present study demonstrates a significant correlation of the nasal and inferior RNFLT with the degree of refractive error. The nasal and inferior RNFLT measured with OCT were thicker in hyperopic eyes. Upon evaluation of the RNFLT of high myopic patients, a mistaken diagnosis of glaucoma should not be made because of a thin nasal and inferior RNFLT as compared with that of emetropic or hyperopic patients.

Additional studies should be conducted to determine the clinical usefulness of OCT in detecting early glaucomatous damage and evaluating the degree of glaucoma. OCT will be a more reliable diagnostic tool once the significant asymmetry of RNFLT between corresponding sectors, divided according to the vertical or horizontal meridian and between the right and left eye, is understood. It is especially notable that a significant asymmetry of nasal and temporal RNFLT between the right and left eye in normal subjects was found. A misdiagnosis for glaucoma should not be made in a situation where a difference of RNFLT in nasal or temporal sectors between the right and left eye is detected.