Optical coherence tomography in the evaluation of structural changes in primary open-angle glaucoma with and without elevated intraocular pressure

Background/Aim. Glaucoma is a progressive optic neuropathy characterized by damage of the retinal ganglion cells and their axons and glial cells. The aim of this study was to evaluate the differences and connections between changes in the visual field and the thickness of the peripapillary retinal nerve fiber layer (RNFL), using optical coherence tomography (OCT) in patients with primary open-angle glaucoma with normal and elevated intraocular pressure (IOP). Methods. This prospective study included 38 patients (38 eyes) with primary open-angle glaucoma with normal intraocular pressure (NTG) and 50 patients (50 eyes) with primary open-angle glaucoma with elevated intraocular pressure (HTG), paired by the same degree of structural glaucomatous changes in the optic nerve head and by age. OCT protocols ’fast RNFL thickness’ and ’fast optic disc’ were used for testing. The patients’ age, gender, best corrected visual acuity (BCVA), IOP, stereometric and functional parameters were compared. Results. The average age of the examined population was 65.49 ± 9.36 (range 44–83) years. There was no statistically significant difference by age and by gender between the two study groups (p = 0.795 and p = 0.807, respectively). BCVA was higher in patients with NTG but there was no statistically significant difference compared to HTG patients (p = 0.160). IOP was statistically significantly higher in patients with HTG compared to NTG patients (17.40 ± 2.77 mmHg vs 14.95 ± 3.01 mmHg, p = 0.009). The cup/disc (C/D) (p = 0.258), mean deviation (MD) (p = 0.477), corrected patern standard deviation (CPSD) (p = 0.943), disk area (p = 0.515), rim area (p = 0.294), rim volume (p = 0.118), C/D area R (p = 0.103), RNFL Average (p = 0.632), RNFL Superior (p = 0.283) and RNFL Inferior (p = 0.488) were not statistically significantly different between the groups. Conclusion. OCT measurements of the RNFL thickness provide clinically significant information in monitoring of glaucomatous changes. There are no differences in the patterns of RNFL defects per sectors and quadrants between NTG and HTG, measured by OCT.

Optical coherence tomography (OCT) is a contactless, non-invasive diagnostic procedure, that can be repeated number of times [23][24][25][26][27] .OCT provides measurements of the thickness of the retinal nerve fiber layer, the loss of 25% of which, occurs in the early stages of glaucoma 4,28 .Some authors have shown that the loss of as much as 40% to 50% of the nerve fibers of the optic nerve does not have to be accompanied by changes in the visual field 28 .Because of this change, RNFL thickness is considered as the most sensitive indicator for glaucomatous damage 29 .
Yamazaki et al. 30 , and Woo et al. 31 demonstrated the different patterns of retinal nerve fiber layer (RNFL) defects in patients with HTG compared to patients with NTG, ie. in NTG the defects are more localized.However, Kubota et al. 32 , and Kook et al. 33 reported that there was a symmetrical loss of RNFL in the superior and inferior quadrants in POAG patients with elevated IOP.In contrast to all this, Mok et al. 1,34 reported that there are no differences in RNFL thinning pattern in HTG and NTG; it was a mixture of localized and diffuse RNFL defects in both cases 29 .
The aim of this study was to evaluate the differences and connections between changes in the visual field and the thickness of the peripapillary retinal nerve fiber layer, measured by OCT in POAG patients with normal and elevated IOP.

Methods
This prospective study included 38 eyes (38 patients) suffering from POAG with normal IOP (NTG) and 50 eyes (50 patients) suffering from POAG with elevated IOP (HTG), paired by the degree of structural glaucomatous changes of the optic nerve head and by age.The selection and monitoring of the patients was performed at the Department of Ophthalmology, University Medical Center Niš.The research followed the Declaration of Helsinki and was approved by the Ethical Review Board of Clinical Center Niš.
All of the patients had the previous diagnosis of POAG with elevated or normal IOP.Exclusion criteria were: spherical equivalent > ± 4 Dsph, other ocular diseases, ocular surgery performed in the previous year, laser treatment, trauma, secondary glaucoma, neurological diseases with changes in the visual field, patients with unreliable visual field (defined as false-negative errors >33%, false-positive errors > 33%, and fixation losses > 20%), patients with anomalies of the optic nerve head.If both eyes of the patients satisfied the inclusion criteria, only one eye was randomly selected.
The following demographic and stereometric parameters were measured and studied here: the patients' age (years); gender; best corrected visual acuity (BCVA) determined by Snellen; IOP (in mmHg) measured by applanation tonometry; the size of the optic nerve head excavation (cup/disc -C/D) determined by indirect ophthalmoscopy with 90 D lens; the value of mean deviation (MD, in dB) and corrected pattern standard deviation (CPSD, in dB) obtained by computed (static) perimetry; disc area (in mm 2 ), rim area (in mm 2 ), rim volume (in mm 3 ), C/D area ratio, the global average thickness of peripapillary retinal nerve fibers (RNFL Avg, in µm) and average RNFL thickness (in µm) in four quadrants and all sectors measured by OCT.
Static perimetry, Humphrey visual field analyzer (Carl Zeiss Meditec, Inc., treshold test 24-2), was used to reveal the defects in the visual field of patients.The first visual field was never considered with regard to the learning curve.
OCT measurements were performed on the Stratus OCT (Carl Zeiss Meditec, Inc.).For, testing, 'fast RNFL thickness' and 'fast optic disc' OCT protocols were used.Fast RNFL protocol consists of three circular scans, each 3.46 mm in diameter, centered on the optic disc.This diameter has been shown to be optimal for RNFL thickness analysis

Results
The study included 38 (43.18%)NTG patients and 50 (56.82%)HTG patients.The average age of the examined population was 65.49 ± 9.36 (range 44-83) years.No statistically significant difference by age and by gender was found between the two study groups (p = 0.795 and p = 0.807, respectively) (Table 1).
Best corrected visual acuity was higher in patients with NTG but there was no statistically significant difference compared to HTG patients (p = 0.160).IOP with the prescribed therapy was significantly higher in patients with HTG compared to NTG patients (17.40 ± 2.77 mmHg vs 14.95 ± 3.01 mmHg, p = 0.009).The morphological parameters of the optic nerve head, due to the examination of the fundus with the 90D lens, C/D, were not statistically significantly different between the patients with HTG and NTG (p = 0.258).Changes in the visual field, ie.functional parameters: MD (p = 0.477) and CPSD (p = 0.943) did not show any statistical significance.Stereometric ONH parameters obtained by OCT: disk area (p = 0.515), rim area (p = 0.294), rim volume (p = 0.118), C/D area R (p = 0.103), RNFL Avg (p = 0.632), RNFL Superior (RNFL Sup) (p = 0.283) and RNFL Inferior (RNFL Inf) (p = 0.488) were not statistically significantly different between the groups (Table 1, Figure 1).The average global and quadrant, and sectoral RNFL thickness values were compared between the groups of NTG and HTG patients; no statistically significant difference in the values of the thickness of RNFL as per quadrant and sectors was found (Tables 1 and 2).
The correlation analysis of the structural and functional parameters in patients with NTG and HTG was performed by Spearman's correlation coefficient (Tables 3 and 4  For explanation see under Table 3.

Discussion
Glaucoma is associated with the loss of retinal ganglion cells, which manifests as defects in the RNFL.The most important part of glaucoma diagnosis is the evaluation of the thickness of the retinal nerve fibers and changes of the optic disc.This can be assessed by OCT, as a good, objective and repeatable method.Also, OCT with its high axial resolution of 9-10 µm is a much superior method than GDx VCC and Heidelberg retina tomograph (HRT) III 1,29 .For a long time it was thought that there are different patterns of RNFL thinning occurring in patients with POAG with elevated IOP and without elevated IOP.Kubota et al. 32 found a difference in RNFL defects in patients with HTG and NTG.RNFL thickness was reduced symmetrically in the superior and inferior quadrants in HTG patients, whereas in patients with NTG a more localized RNFL loss in inferior quadrant occurs 32,35 .Quite different results have appeared in recent years: there are no differences in RNFL thinning pattern between HTG and NTG, measured by OCT 29 .Our research also shows no differences in the defect pattern (localized or diffuse) of RNFL in HTG and NTG patients, and no differences in the amount of lost retinal nerve fibers of the two groups of patients.The strength of this statement is supplemented by the fact that the patients were at the same disease stage, with the same degree of structural damage of the optic nerve head and the same degree of change in the visual field (MD < 12 dB).In addition, all glaucoma patients had spherical equivalent (SE) within ± 4 Dsph, so the possibility of involving patients with very large or small optic nerve head (ONH) was reduced to a minimum.Our results are consistent with Mok et al. 1,34 , and Konstantakopoulou et al. 29 published results, that patients with HTG and NTG showed a similar RNFL thinning pattern, which is a mixture of diffuse and localized changes, measured by OCT 36 .
Studying the localized RNFL loss by clock-hour sectors did not show any statistically significant difference, although this method provides more precise information on smaller fields of peripapillary nerve fiber loss.Furthermore, OCT paremeters for inferotemporal and superotemporal sector, which are considered to be of great help for the early diagnosis of glaucoma and making a difference compared to healthy eyes, did not differ significantly between our groups of HTG and NTG patients 1,37 .The study did not include pa-tients with ocular hypertension as well as a healthy population because the differences between these two groups of patients with POAG patients are already tested and proven.
The results from this study related to age, gender, BCVA and morphological parameters of the optic nerve head did not differ significantly between the patients with HTG and NTG, and are fully in line with other publications 2,29,34,38 .Only the IOP was significantly lower in patients with NTG (p = 0.009) 2 .Häntzschel et al. 2 reported that visual acuity was significantly better in patients with NTG than in patients with HTG (p < 0.002); we found that BCVA was higher in patients with NTG but there was no statistically significant difference compared to HTG patients (p = 0.160), most likely due to better randomization.The sample size for this study is relatively small (38 NTG, 50 HTG), but it is in accordance with the groups studied by other authors: Mok et al. 34  (38 NTG, 48 HTG), and Konstantakopoulou et al. 29  (20 NTG, 21 HTG).
Sihota et al. 28 found that defects in the visual field (MD) positively correlate with the thickness of RNFL superior, inferior, nasal, temporal and average, and that there is a negative correlation between RNFL (average, superior, nasal, inferior, temporal) parameters with CPSD.Shin et al. 38 reported the connection between MD and CPSD and RNFL Avg and RNFL Inf thickness parameters.Our results of two visual field indices (MD and CPSD) showed that there is a statistically significant positive correlation between RNFL Temp and MD (ρ = 0.454, p = 0.022) and negative correlation between RNFL Temp and CPSD (ρ = -0.470,p = 0.018), only in patients with HTG.This difference could be due to the difference in sample size and because the other authors compared the patients with POAG and healthy population 28 .

Conclusion
OCT measurements of RNFL thickness provide clinically significant information in the monitoring of glaucomatous changes.There are no differences in RNFL thinning pattern per sectors and quadrants between NTG and HTG, as assessed by OCT.However, further larger studies are needed to supplement these findings view of the fact that it is known that RNFL thickness measurements and morphology of ONH have wide inter-individual and interracial variations.

Table 1 Demographic and stereometric parameters of primary open-angle glaucoma (POAG) of the patients with normal intraocular pressure (NTG) and elevated intraocular pressure (HTG)
The comparison of the values between patients with NTG and HTG was carried out using t-test or Mann-Whiteney test.The association of these parameters was tested by Spearman's correlation coefficient.Statistical analysis of data was performed by SPSS (version 16.0, SPSS Science, Chicago, IL, USA).p-value < 0.05 was considered statistically significant.