March 2003
Volume 44, Issue 3
Free
Glaucoma  |   March 2003
Reference Plane Definition and Reproducibility in Optic Nerve Head Images
Author Affiliations
  • James C. H. Tan
    From the Glaucoma Research Unit, Moorfields Eye Hospital, London, United Kingdom.
  • Roger A. Hitchings
    From the Glaucoma Research Unit, Moorfields Eye Hospital, London, United Kingdom.
Investigative Ophthalmology & Visual Science March 2003, Vol.44, 1132-1137. doi:10.1167/iovs.02-0039
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      James C. H. Tan, Roger A. Hitchings; Reference Plane Definition and Reproducibility in Optic Nerve Head Images. Invest. Ophthalmol. Vis. Sci. 2003;44(3):1132-1137. doi: 10.1167/iovs.02-0039.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To describe and evaluate a new experimental reference plane for measuring rim area in scanning laser tomography.

methods. The experimental reference plane was positioned so that (1) it always lay entirely below the margin of the optic nerve head (ONH), (2) it remained at a set z-axis distance below the ONH in images of each eye, and (3) it was at a level where variability in rim area is least. Twenty normal control subjects and 20 patients with glaucoma underwent test-retest scanning laser tomographic imaging by same and different operators during same and separate visits. Control subjects had image series spanning at least 3 years. The effect of the positioning of the reference plane on global and regional rim area variability was assessed in intra- and intervisit test-retest images and longitudinal image series and compared with the standard and 320-μm reference planes.

results. Variability in the experimental reference plane was less in test-retest images and longitudinal data (P < 0.05) and more uniform around the ONH than with other reference planes. Variability in the former was not appreciably affected by testing involving different operators and visits, or by the presence of glaucoma.

conclusions. Variability in rim area by the experimental reference plane was significantly less, more uniform around the ONH, not affected by different operators and visits, and less affected by glaucomatous morphology than other reference planes. This difference was pronounced in sequential data and has implications for detecting progression of glaucoma.

Sequential imaging of the optic nerve head (ONH) by scanning laser tomography has been proposed to evaluate progression of glaucoma. 1 2 3 Topography can be analyzed by a reference plane so that parameters such as rim area can be measured and serve as markers of progression. Measurement of parameters relies heavily on this reference plane and it is vital that its position remains unchanged in relation to the tissue of interest. 
Many different reference surfaces have been proposed, but it is not clear which is most useful for assessing progression. Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Heidelberg, Germany) software previously used a “curved” reference surface with a shape that varied with the height profile of the margin of the ONH. 4 Flat reference planes have since been used 5 and various landmarks proposed for their positioning, such as peripapillary sclera, 6 7 retinal pigment epithelium, 8 peripapillary retina, 9 10 and retinal height at the margin of the ONH. 5 9 However, it is unclear which reference plane is useful in longitudinal analysis, and hence the reproducibility of different reference planes requires study. 
We have found that rim area variability differs with reference plane positioning 11 and can be explained by fluctuation in height between the ONH and reference plane. 12 To optimize reproducibility, we devised an experimental reference plane that maintains an unchanging height relationship with the ONH throughout an imaging series. The reproducibility of this reference plane was evaluated and compared with the HRT’s standard reference plane and 320-μm reference plane. 
Methods
Reference Plane Analysis
Figure 1 illustrates the different reference plane concepts compared in this study. The standard and 320-μm reference planes have been used widely in studies. 
New Experimental Reference Plane.
For each eye, a mean topography image was selected to be the baseline image in which the position of the reference plane was determined, calculating the position as follows: 
LOW5%.
The contour line’s lowest region was calculated (LOW5%). Heights 1° apart (360 data points) on the contour lines of single topography images (used to derive the baseline mean topography image) were read from the HRT software. Heights were ranked in each single topography image, from which the mean of the lowest 5% of heights was calculated. The means of the lowest 5% of heights for each of the three single topography images were averaged to arrive at LOW5%
R.
The reference plane was positioned to ensure that it lay beneath the entire circumference of the contour line. A reference plane lying above part of the contour line underestimates adjacent rim tissue. To determine R, variability was analyzed in longitudinal image series of normal control eyes. R was the distance of the reference plane beneath LOW5%, where rim area variability was least. This was held constant once calculated. 
REFdis.
Once positioned at R, the z-axis distance of the reference plane below mean height of the contour line (MHC; which is the mean height of locations on the contour line measured in relation to the reference ring) was calculated for each ONH and called REFdis. MHC was used as a marker of the topographical z-axis position of the ONH. REFdis is unique to each ONH and, once calculated in the baseline image, is kept constant in all the images of an eye. By keeping REFdis constant, we sought to keep the height relationship between the ONH and reference plane constant in image series of the same eye. REFdis can be expressed as:  
\[\mathrm{REF}_{\mathrm{dis}}\ {=}\ \mathrm{LOW}_{5\%}\ {+}\ \mathrm{R}\]
 
REFpos.
Position of the reference plane, or REFpos, can thus be expressed as:  
\[\mathrm{REF}_{\mathrm{pos}}\ {=}\ \mathrm{MHC}\ {+}\ \mathrm{REF}_{\mathrm{dis}}\]
 
\[{=}\ \mathrm{MHC}\ {+}\ \mathrm{LOW}_{5\%}\ {+}\ \mathrm{R}\]
 
Standard Reference Plane.
This plane was set 50 μm posterior to the mean contour line height between 350° and 356° on the contour line (HRT software versions 1.11–2.01, and HRT II). 5 9  
The 320-μm Reference Plane.
This plane was offset 320 μm posterior to the mean height of the “reference ring” (HRT software versions 1.09–1.10). The reference ring is centered on the image frame and located in its periphery, had an outer diameter 94% and a width 3% of the image size, and is the zero-referencing region for pixel height. 9  
Subjects and Imaging
Variability in rim area was studied in test-retest images of age-matched subjects with glaucoma and normal control subjects, and in longitudinal series of images of normal eyes. Normal subjects and those with glaucoma attended the Ocular Hypertension and Early Glaucoma Research Clinic at Moorfields Eye Hospital and underwent the same protocol of repeat testing. All had previously undergone scanning laser tomography and perimetry at least six times. Table 1 shows subjects’ demographic information. This study adhered to the tenets of the Declaration of Helsinki and had appropriate Institutional Review Board approval and the subjects’ informed consent. 
Normal subjects were volunteers comprising spouses or friends of hospital patients, hospital staff, or members of external nonmedical social organizations. They had IOP persistently less than 22 mm Hg; normal and reliable serial visual fields (Humphrey 24-2; Humphrey Systems, Dublin, CA), with an Advanced Glaucoma Intervention Study (AGIS) 13 score of 0; no concurrent ocular disease; no family history of glaucoma; and refractive errors less than ±6 D. All were more than 40 years of age. The appearance of the ONH was not taken into account for entry into the study. Patients with glaucoma had had pretreatment IOP higher than 21 mm Hg on at least two occasions; reproducible field defects (24-2 program; Humphrey Systems) with AGIS scores higher than 0; open anterior chamber angles; and no known ocular disease other than glaucoma. All patients were treated medically and had IOP lower than 22 mm Hg at the time of testing. Neither ONH appearance nor severity of visual field abnormality was used to restrict entry into the study. 
Test-retest scanning laser tomography (HRT software ver. 2.01; Heidelberg Engineering) of both eyes of 20 normal and 20 age-matched subjects with glaucoma was conducted by experienced operators. Three well-centered 10° single topography images were acquired at each session. Corneal curvature, scan depth, and focus settings were kept constant. Pupils were not dilated. Normal subjects attended two test visits separated by 6 to 8 months. Each visit comprised two imaging sessions separated by at least 1 hour. The same operator performed the scans in both imaging sessions of the first visit and in one session of the second visit. A second operator performed the scans in the other imaging sessions of the second visit. Patients with glaucoma underwent test-retest imaging in only one visit, performed by the same operator. Imaging sessions were randomly scheduled. Test-retest variability was analyzed for testing by: the same operator in the same visit (same operator-same visit) and by different operators in separate visits (different-operator-different-visit). This is relevant because longitudinal imaging in chronic glaucoma is likely to be performed in many different visits over time by different operators. 
Nineteen normal subjects had longitudinal image series from at least six imaging visits over at least 3 years. Different operators with various levels of experience had acquired these images because of the turnover of research clinic technical staff. Each series was analyzed separately. 
Image Analysis and Statistics
Mean topography images from one randomly selected eye of subjects in the glaucoma and normal groups were analyzed by the three reference planes. Mean images were generated from triplets of single topography images and used if mean pixel standard deviation was less than 50 μm, with pixel height measured by the reference ring. A contour line was outlined corresponding to the inner margin of the scleral ring of Elschnig on a mean topographic image of each subject (all performed by JCHT). Stereoscopic optic disc photographs were referred to if needed. Contour lines were then exported to related test-retest mean topography images. Rim area was evaluated globally and regionally. Patterns of variability around the ONH were examined by assessing regional rim area in 30° sectors around the ONH (0–360°). 
Variability in test-retest images was represented by agreement 14 between pairs of images, expressed in graphs as the width of agreement intervals (95% confidence intervals of differences). For regional variability, analysis was modified so that agreement intervals for rim area sectors were presented in bar graphs by angle (0–360°). 
Variability in longitudinal image series was represented by standard deviation, which was used to estimate variability so that R could be derived and in longitudinal data. The width of the intervals for the 5th to 95th percentiles and median standard deviations for each sector were plotted in bar graphs. Significance testing in nonparametric data was by the Wilcoxon matched-pairs test (signed rank sum test) for comparing paired data. Statistical analysis was conducted by computer (SPSS ver. 9.0 for Windows; SPSS Inc, Chicago, IL). 
Results
Derivation of R
Figure 2 shows a trend toward decreasing rim area variability as the reference plane descended from 0 to 100 μm below LOW5%. Variability was least between 100 and 120 μm but increased when the reference plane was more than 120 μm below LOW5%. R of 100 μm was chosen for subsequent analysis. 
Variability in Rim Area in Test-Retest Data
Figure 3 shows that 95% confidence intervals of differences (agreement intervals) for global variability in normal eyes were narrower with the experimental reference plane than standard and 320-μm reference planes. All agreement intervals tended to be wider with different-operator-different-visit testing than same-operator-same-visit testing. This tendency was less, however, in experimental reference plane data than data from other reference planes: widening of agreement intervals with the experimental reference plane was 23% and 17% less than with the standard and 320-μm reference planes, respectively. 
Figure 4 shows polar plots of agreement intervals for regional rim area variability in normal and glaucomatous eyes. In normal control subjects, agreement intervals tended to be wider temporally (0–90° and 270–360°) than nasally in all reference planes. In glaucomatous eyes, agreement intervals tended to be widest temporally with the experimental and 320-μm reference planes but widest nasally with the standard reference plane. Same-operator-same-visit regional variability was more uniform around the ONH with the experimental reference plane (left) than with other reference planes (middle and right). Unlike other reference planes, the variability profile of experimental reference plane data did not change appreciably with different-operator-different-visit testing or in testing glaucomatous eyes. 
Variability in Rim Area in Longitudinal Series of Images
Figure 5 shows that the patterns of regional variability differed between reference planes. The 5th- to 95th-percentile intervals of sector variability were wider in standard and 320-μm reference plane data than in experimental reference plane data for every sector. Agreement intervals for the standard and 320-μm reference planes were widest temporally and nasally. Intervals in the former were particularly wide nasally and in keeping with findings in standard reference plane test-retest data. Again, experimental reference plane variability was more uniform around the ONH than in standard and 320-μm reference plane data. 
Figure 6 shows rim area sectors in longitudinal image series with significantly less variability (P < 0.05), when measured by the experimental reference plane compared with the standard (Fig. 6 , left) and 320-μm (Fig. 6 , right) reference planes. The pattern of sector variability differed between standard and 320-μm reference plane data. 
Discussion
Increased variability in rim area can be expected if the ONH and reference plane shift in relation to each other from image to image. 12 Analyzed by the experimental reference plane, rim area variability tended to be less than in standard or 320-μm reference plane analysis. Variability around the ONH was also more uniform, was not appreciably affected by glaucomatous morphology or different test conditions, and was significantly (P < 0.05) less than other reference planes in longitudinal analysis. 
No two ONHs are identical, and thus analysis should adapt to suit morphologic variations. The experimental reference plane was devised to lie entirely beneath the margin of the ONH by factoring the height profile of each contour line into calculations. Figure 7 illustrates why this is necessary. LOW5% was the level in each ONH above which the reference plane should not rise. The value for LOW5% was not preselected but was derived from the lowest 5% of contour line heights of each ONH and averaged from multiple topographies to minimize the effect of randomly outlying values. R of 100 μm (100 μm below LOW5%) was relatively deep and compatible with least variability. With this R, the reference plane is more likely to remain entirely below the ONH margin despite topographical variability. The cup is probably steeper here as well. Rim area is affected more when the reference plane shifts on a gradually sloping than a steep cup, as suggested by Figures 1 and 7 . With R deeper than 120 μm, variability tended to increase, possibly because of proximity to major vascular trunks and the lamina cribrosa, especially in shallow cups. REFpos was derived from all heights on the contour line so that the reference plane’s position would not rely on localized regions or presume selectivity in the pattern of change. 
Unlike other definitions, the experimental reference plane compensates for positional changes between the reference plane and ONH. This is the main reason for its reduced variability. 12 15 16 Testing involving different operators and visits probably exaggerates these positional changes. The 320-μm reference plane does not cater well to morphologic variations and may lie above parts of the ONH margin, 5 as depicted in Figure 7 . The standard reference plane is superficial in the temporal ONH (50 μm below the surface), where cup slope is gradual and susceptible to variability if the reference plane shifts. In glaucoma, the nasal ONH surface may be depressed and come to lie beneath the standard reference plane, which is fixed temporally. Because the standard reference plane is positioned by a small 6° section of the inferotemporal ONH where it pivots, tilting of the image or shifts in reference plane may cause variability opposite the pivot nasally. 
Reference planes have usually been positioned by landmarks presumed to be relatively unaffected by glaucoma, although whether this is true has been debated. 17 The experimental reference plane differs in being positioned by a landmark that is expected to change in glaucoma. But there are reasons that this position is useful. First, experimental reference-plane data were more reproducible than other reference planes and not appreciably influenced by varied test conditions or glaucomatous morphology. Variability that changes with repeat testing and advancing damage can confound analysis. Second, the experimental reference plane compensated for shifts of the ONH that can affect analysis were the reference plane not to shift in tandem, as illustrated in Figure 8 . Changed rim area should reflect disease rather than such shifts. Third, positioning the reference plane at the same distance below the ONH surface allowed standardized measurement of rim area in imaging series. The ONH may shift in relation to external landmarks, such as the reference ring, which if used to anchor the reference plane can result in measurement artifact (Fig. 8) . Bias in measuring height and volume by the experimental reference plane can be expected, but this is less so with rim area because it is estimated perpendicular to the z-axis. Fourth, detecting progression depends on the system’s signal-to-noise ratio. The z-axis position of the ONH fluctuated considerably in image series: 50% had a range of MHC exceeding 120 μm, with this range greater than 200 μm in some. Normal human retinal nerve fiber layer (RNFL) at the ONH margin is 316 to 406 μm thick, 18 and so MHC variability is considerable in relation to normal RNFL thickness. MHC variability may even exceed RNFL thickness in glaucoma if significant axons are already lost by the time visual field defects appear. Empiric data suggest that the benefit of significantly reduced variability should outweigh the possibility that MHC is affected by glaucoma. Figure 8 shows that experimental reference plane analysis is sensitive to subtle rim area progression despite markedly reduced MHC. 
Variability in rim area was less and more uniform around the ONH when the z-axis distance between the experimental reference plane and ONH was held constant. More uniform variability should be simpler to account for and help limit bias in detecting localized progression. Application of the experimental reference plane in progressive glaucoma is outside the scope of the present study but will be the subject of a future study. 
 
Figure 1.
 
Different positions of reference planes. Top: experimental reference plane; middle: standard reference plane; bottom: 320-μm reference plane.
Figure 1.
 
Different positions of reference planes. Top: experimental reference plane; middle: standard reference plane; bottom: 320-μm reference plane.
Table 1.
 
Demographic Data for Normal and Glaucoma Groups
Table 1.
 
Demographic Data for Normal and Glaucoma Groups
Normal Glaucoma P
20 20
Age (y) 66.5 ± 9.2 68.8 ± 11.5 0.70
Male 12 10
Right eyes 13 10
Visual field MD (dB) +0.11 ± 1.2 −4.6 ± 1.9 0.0000
Visual field CPSD (dB) 0.89 ± 0.75 3.3 ± 1.8 0.0000
Mean SD (μm) of mean images 22.6 ± 8.1* 25.9 ± 9.65* 0.17*
22.9 ± 10.2 27.2 ± 10.5 0.16
Figure 2.
 
Variability in rim area with the reference plane at different z-axis distances below LOW5%.
Figure 2.
 
Variability in rim area with the reference plane at different z-axis distances below LOW5%.
Figure 3.
 
Global rim area variability in normal eyes measured by three different reference planes. (•) Experimental reference plane; (▪) standard reference plane; (▴) 320-μm reference plane; left, same-operator-same-visit testing; right, different-operator-different-visit testing. In each graph: middle horizontal lines: mean difference, flanking horizontal lines: 95% confidence intervals of differences.
Figure 3.
 
Global rim area variability in normal eyes measured by three different reference planes. (•) Experimental reference plane; (▪) standard reference plane; (▴) 320-μm reference plane; left, same-operator-same-visit testing; right, different-operator-different-visit testing. In each graph: middle horizontal lines: mean difference, flanking horizontal lines: 95% confidence intervals of differences.
Figure 4.
 
Polar plots for variability in sectors of rim area. y-Axis: 95% confidence intervals of differences; circumference, angular location of 30° sectors around the ONH, where 0° is temporal, 90° is superior, 180° is nasal, and 270° is inferior; center point: variability = 0. Left: experimental reference plane; middle: standard reference plane; right: 320-μm reference plane. Bold line: same-operator-same-visit testing in normal eyes; (•) different-operator-different-visit testing in normal eyes; (▴) same-operator-same-visit testing in glaucomatous eyes.
Figure 4.
 
Polar plots for variability in sectors of rim area. y-Axis: 95% confidence intervals of differences; circumference, angular location of 30° sectors around the ONH, where 0° is temporal, 90° is superior, 180° is nasal, and 270° is inferior; center point: variability = 0. Left: experimental reference plane; middle: standard reference plane; right: 320-μm reference plane. Bold line: same-operator-same-visit testing in normal eyes; (•) different-operator-different-visit testing in normal eyes; (▴) same-operator-same-visit testing in glaucomatous eyes.
Figure 5.
 
Polar plots for sector rim area variability in the longitudinal data of normal eyes. Bold: experimental reference plane; circles: standard reference plane; triangles: 320-μm reference plane.
Figure 5.
 
Polar plots for sector rim area variability in the longitudinal data of normal eyes. Bold: experimental reference plane; circles: standard reference plane; triangles: 320-μm reference plane.
Figure 6.
 
Sectors where rim area variability was significantly less in experimental reference plane data compared with the standard (left) and 320-μm (right) reference plane data. Shaded sectors: P < 0.05; unshaded sectors: P > 0.05. Sector locations: 0°, temporal; 90°, superior; 180°, nasal; 270°, inferior.
Figure 6.
 
Sectors where rim area variability was significantly less in experimental reference plane data compared with the standard (left) and 320-μm (right) reference plane data. Shaded sectors: P < 0.05; unshaded sectors: P > 0.05. Sector locations: 0°, temporal; 90°, superior; 180°, nasal; 270°, inferior.
Figure 7.
 
Description of rim area by the standard (top left) and 320-μm (bottom left) reference planes in the same image. Adjacent to the images are renderings of the ONH in cross-section: profiles: height along the contour line marking the ONH margin; top horizontal line: MHC; bottom horizontal line: reference plane. With the standard reference plane (top), rim area is represented as expected. With the 320-μm reference plane (bottom), the plane is situated above the temporal margin of the ONH, causing rim area to be grossly underestimated. The variation in rim area due to different reference plane positions is not uniform around the ONH but much greater temporally than nasally. This is probably because the surface of nasal rim is relatively high (profiles between 90° and 270°) with its adjacent cup walls steeper.
Figure 7.
 
Description of rim area by the standard (top left) and 320-μm (bottom left) reference planes in the same image. Adjacent to the images are renderings of the ONH in cross-section: profiles: height along the contour line marking the ONH margin; top horizontal line: MHC; bottom horizontal line: reference plane. With the standard reference plane (top), rim area is represented as expected. With the 320-μm reference plane (bottom), the plane is situated above the temporal margin of the ONH, causing rim area to be grossly underestimated. The variation in rim area due to different reference plane positions is not uniform around the ONH but much greater temporally than nasally. This is probably because the surface of nasal rim is relatively high (profiles between 90° and 270°) with its adjacent cup walls steeper.
Figure 8.
 
Illustration of how reference planes differ in determining rim area. ONH images (A) were acquired over the period when a visual field defect was developing in a subject with ocular hypertension; (B) experimental reference plane; (C) standard reference plane; (D) 320-μm reference plane, left: t = 0, right: t = 3 years. (A) Change was apparent as subtle vertical elongation of the cup affecting the inferior pole (thin arrows). Open arrow: prominence of the vessel trunk, which serves as a marker for z-axis position. MHC: 3 μm (left) and 139 μm (right), representing a height reduction of 136 μm, which could be caused by glaucoma, although the effect of variability cannot be excluded. In (B) the rendering of rim area closely resembles the appearance of the rim in (A). Despite markedly reduced MHC, inferior elongation of the cup is evident from left to right, with the reference plane retaining a similar level in the ONH (vascular trunk prominence represented similarly). Hence, the ONH and reference plane do not seem to have shifted in relation to each other. In (C), the reference plane is superficial to the vascular trunk (not seen) and the cup is wider. The reference plane in the right image appears to lie even more superficially than in the left image (to the vascular prominence) and has a wider cup. Rim appearance in (C) does not closely resemble that of (A). In (D, left), the z-axis position of the reference plane in relation to the vascular prominence is similar to that of (B). In the right image, however, the large reduction of MHC causes the reference plane to lie more superficially in the ONH (above the vessel trunk), rendering the cup diffusely larger and appearing different from (A, right).
Figure 8.
 
Illustration of how reference planes differ in determining rim area. ONH images (A) were acquired over the period when a visual field defect was developing in a subject with ocular hypertension; (B) experimental reference plane; (C) standard reference plane; (D) 320-μm reference plane, left: t = 0, right: t = 3 years. (A) Change was apparent as subtle vertical elongation of the cup affecting the inferior pole (thin arrows). Open arrow: prominence of the vessel trunk, which serves as a marker for z-axis position. MHC: 3 μm (left) and 139 μm (right), representing a height reduction of 136 μm, which could be caused by glaucoma, although the effect of variability cannot be excluded. In (B) the rendering of rim area closely resembles the appearance of the rim in (A). Despite markedly reduced MHC, inferior elongation of the cup is evident from left to right, with the reference plane retaining a similar level in the ONH (vascular trunk prominence represented similarly). Hence, the ONH and reference plane do not seem to have shifted in relation to each other. In (C), the reference plane is superficial to the vascular trunk (not seen) and the cup is wider. The reference plane in the right image appears to lie even more superficially than in the left image (to the vascular prominence) and has a wider cup. Rim appearance in (C) does not closely resemble that of (A). In (D, left), the z-axis position of the reference plane in relation to the vascular prominence is similar to that of (B). In the right image, however, the large reduction of MHC causes the reference plane to lie more superficially in the ONH (above the vessel trunk), rendering the cup diffusely larger and appearing different from (A, right).
Kamal, DS, Viswanathan, AC, Garway-Heath, DF, Hitchings, RA, Poinoosawmy, D, Bunce, C. (1999) Detection of optic disc change with the Heidelberg retina tomograph before confirmed visual field change in ocular hypertensives converting to early glaucoma Br J Ophthalmol 83,290-294 [CrossRef] [PubMed]
Chauhan, BC, Blanchard, JW, Hamilton, DC, LeBlanc, RP. (2000) Technique for detecting serial topographic changes in the optic disc and peripapillary retina using scanning laser tomography Invest Ophthalmol Vis Sci 41,775-782 [PubMed]
Chauhan, BC, McCormick, TA, Nicolela, MT, Le Blanc, RP. (2001) Optic disc and visual field changes in a prospective longitudinal study of patients with glaucoma: comparison of scanning laser tomography with conventional perimetry and optic disc photography Arch Ophthalmol 119,1492-1499 [CrossRef] [PubMed]
Burk, ROW, Rohrschnider, K, Volcker, HE, Zinser, G. (1990) Analysis of three-dimensional optic disc topography by scanning laser tomography Nasemann, JE Burk, ROW eds. Laser Scanning Ophthalmoscopy and Tomography ,161-171 Quintessenz Munich, Germany.
Burk, R, Vihanninjoki, K, Bartke, T, et al (2000) Development of the standard reference plane for the Heidelberg Retina Tomograph Graefes Arch Clin Exp Ophthalmol 238,375-384 [CrossRef] [PubMed]
Maier, H, Siebert, M, Gramer, E, Kampik, A. (1990) Eine Mebsszahl für die nervenfaserschichtdicke: messungen mit dem laser tomographic scanner (LTS) Gramer, E eds. Glaukom: diagnostic und therapie ,120-145 Enke Stuttgart, Germany.
Frohn, A, Jean, B, Zinser, G, Thiel, HJ. (1990) The problem of reference plane definition for cup volume measurements Nasemann, JE Burk, ROW eds. Laser Scanning Ophthalmoscopy and Tomography ,197-206 Quintessenz Munich, Germany.
Tuulonen, A, Vihanninjoki, K, Airaksinen, PJ, Alanko, H, Nieminen, H. (1993) The effect of reference levels on neuroretinal rim area and rim volume measurements in the Heidelberg Retina Tomograph (HRT) [ARVO Abstract] Invest Ophthalmol Vis Sci 35(4),S1729Abstract nr 2209
Operation Manual for the Heidelberg Retina Tomograph Software. Versions 1.09–2.01 1993–1999 Heidelberg Engineering Heidelberg, Germany.
Miller, JM, Caprioli, J. (1992) An optimal reference plane to detect glaucomatous nerve fiber layer abnormalities with computerized image analysis Graefes Arch Clin Exp Ophthalmol 230,124-128 [CrossRef] [PubMed]
Rai, P, Tan, JCH, Hitchings, RA. (2001) Variability of topographic measurements across regions of the optic nerve head [ARVO Abstract] Invest Ophthalmol Vis Sci 42(4),S133Abstract nr 704
Sharma, K, Tan, JCH, Hitchings, RA. (2001) Sources of variability in three-dimensional optic nerve head quantitative analysis [ARVO Abstract] Invest Ophthalmol Vis Sci 42(4),S134Abstract nr 715
. The AGIS investigators (1994) Advanced Glaucoma Intervention Study 2. Visual field test scoring and reliability Ophthalmology 101,1445-1455 [CrossRef] [PubMed]
Bland, JM, Altman, DG. (1986) Statistical methods for assessing agreement between two methods of clinical measurement Lancet 1,307-310 [PubMed]
Menezes, AV, Giunta, M, Chisholm, L, et al (1995) Reproducibility of topographic measurements of the macula with the scanning laser ophthalmoscope Ophthalmology 102,230-235 [CrossRef] [PubMed]
Ang, A, Tong, L, Vernon, SA. (2000) Improvement of reproducibility of macular volume measurements using the Heidelberg retinal tomograph Br J Ophthalmol 84,1194-1197 [CrossRef] [PubMed]
Chen, E, Gedda, U, Landau, I. (2001) Thinning of the papillomacular bundle in the glaucomatous eye and its influence on the reference plane of the Heidelberg retinal tomograph J Glaucoma 10,386-389 [CrossRef] [PubMed]
Varma, R, Skaf, M, Barron, E. (1996) Retinal nerve fiber layer thickness in normal human eyes [published correction appears in Ophthalmology. 1997;104:174] Ophthalmology 103,2114-2119 [CrossRef] [PubMed]
Figure 1.
 
Different positions of reference planes. Top: experimental reference plane; middle: standard reference plane; bottom: 320-μm reference plane.
Figure 1.
 
Different positions of reference planes. Top: experimental reference plane; middle: standard reference plane; bottom: 320-μm reference plane.
Figure 2.
 
Variability in rim area with the reference plane at different z-axis distances below LOW5%.
Figure 2.
 
Variability in rim area with the reference plane at different z-axis distances below LOW5%.
Figure 3.
 
Global rim area variability in normal eyes measured by three different reference planes. (•) Experimental reference plane; (▪) standard reference plane; (▴) 320-μm reference plane; left, same-operator-same-visit testing; right, different-operator-different-visit testing. In each graph: middle horizontal lines: mean difference, flanking horizontal lines: 95% confidence intervals of differences.
Figure 3.
 
Global rim area variability in normal eyes measured by three different reference planes. (•) Experimental reference plane; (▪) standard reference plane; (▴) 320-μm reference plane; left, same-operator-same-visit testing; right, different-operator-different-visit testing. In each graph: middle horizontal lines: mean difference, flanking horizontal lines: 95% confidence intervals of differences.
Figure 4.
 
Polar plots for variability in sectors of rim area. y-Axis: 95% confidence intervals of differences; circumference, angular location of 30° sectors around the ONH, where 0° is temporal, 90° is superior, 180° is nasal, and 270° is inferior; center point: variability = 0. Left: experimental reference plane; middle: standard reference plane; right: 320-μm reference plane. Bold line: same-operator-same-visit testing in normal eyes; (•) different-operator-different-visit testing in normal eyes; (▴) same-operator-same-visit testing in glaucomatous eyes.
Figure 4.
 
Polar plots for variability in sectors of rim area. y-Axis: 95% confidence intervals of differences; circumference, angular location of 30° sectors around the ONH, where 0° is temporal, 90° is superior, 180° is nasal, and 270° is inferior; center point: variability = 0. Left: experimental reference plane; middle: standard reference plane; right: 320-μm reference plane. Bold line: same-operator-same-visit testing in normal eyes; (•) different-operator-different-visit testing in normal eyes; (▴) same-operator-same-visit testing in glaucomatous eyes.
Figure 5.
 
Polar plots for sector rim area variability in the longitudinal data of normal eyes. Bold: experimental reference plane; circles: standard reference plane; triangles: 320-μm reference plane.
Figure 5.
 
Polar plots for sector rim area variability in the longitudinal data of normal eyes. Bold: experimental reference plane; circles: standard reference plane; triangles: 320-μm reference plane.
Figure 6.
 
Sectors where rim area variability was significantly less in experimental reference plane data compared with the standard (left) and 320-μm (right) reference plane data. Shaded sectors: P < 0.05; unshaded sectors: P > 0.05. Sector locations: 0°, temporal; 90°, superior; 180°, nasal; 270°, inferior.
Figure 6.
 
Sectors where rim area variability was significantly less in experimental reference plane data compared with the standard (left) and 320-μm (right) reference plane data. Shaded sectors: P < 0.05; unshaded sectors: P > 0.05. Sector locations: 0°, temporal; 90°, superior; 180°, nasal; 270°, inferior.
Figure 7.
 
Description of rim area by the standard (top left) and 320-μm (bottom left) reference planes in the same image. Adjacent to the images are renderings of the ONH in cross-section: profiles: height along the contour line marking the ONH margin; top horizontal line: MHC; bottom horizontal line: reference plane. With the standard reference plane (top), rim area is represented as expected. With the 320-μm reference plane (bottom), the plane is situated above the temporal margin of the ONH, causing rim area to be grossly underestimated. The variation in rim area due to different reference plane positions is not uniform around the ONH but much greater temporally than nasally. This is probably because the surface of nasal rim is relatively high (profiles between 90° and 270°) with its adjacent cup walls steeper.
Figure 7.
 
Description of rim area by the standard (top left) and 320-μm (bottom left) reference planes in the same image. Adjacent to the images are renderings of the ONH in cross-section: profiles: height along the contour line marking the ONH margin; top horizontal line: MHC; bottom horizontal line: reference plane. With the standard reference plane (top), rim area is represented as expected. With the 320-μm reference plane (bottom), the plane is situated above the temporal margin of the ONH, causing rim area to be grossly underestimated. The variation in rim area due to different reference plane positions is not uniform around the ONH but much greater temporally than nasally. This is probably because the surface of nasal rim is relatively high (profiles between 90° and 270°) with its adjacent cup walls steeper.
Figure 8.
 
Illustration of how reference planes differ in determining rim area. ONH images (A) were acquired over the period when a visual field defect was developing in a subject with ocular hypertension; (B) experimental reference plane; (C) standard reference plane; (D) 320-μm reference plane, left: t = 0, right: t = 3 years. (A) Change was apparent as subtle vertical elongation of the cup affecting the inferior pole (thin arrows). Open arrow: prominence of the vessel trunk, which serves as a marker for z-axis position. MHC: 3 μm (left) and 139 μm (right), representing a height reduction of 136 μm, which could be caused by glaucoma, although the effect of variability cannot be excluded. In (B) the rendering of rim area closely resembles the appearance of the rim in (A). Despite markedly reduced MHC, inferior elongation of the cup is evident from left to right, with the reference plane retaining a similar level in the ONH (vascular trunk prominence represented similarly). Hence, the ONH and reference plane do not seem to have shifted in relation to each other. In (C), the reference plane is superficial to the vascular trunk (not seen) and the cup is wider. The reference plane in the right image appears to lie even more superficially than in the left image (to the vascular prominence) and has a wider cup. Rim appearance in (C) does not closely resemble that of (A). In (D, left), the z-axis position of the reference plane in relation to the vascular prominence is similar to that of (B). In the right image, however, the large reduction of MHC causes the reference plane to lie more superficially in the ONH (above the vessel trunk), rendering the cup diffusely larger and appearing different from (A, right).
Figure 8.
 
Illustration of how reference planes differ in determining rim area. ONH images (A) were acquired over the period when a visual field defect was developing in a subject with ocular hypertension; (B) experimental reference plane; (C) standard reference plane; (D) 320-μm reference plane, left: t = 0, right: t = 3 years. (A) Change was apparent as subtle vertical elongation of the cup affecting the inferior pole (thin arrows). Open arrow: prominence of the vessel trunk, which serves as a marker for z-axis position. MHC: 3 μm (left) and 139 μm (right), representing a height reduction of 136 μm, which could be caused by glaucoma, although the effect of variability cannot be excluded. In (B) the rendering of rim area closely resembles the appearance of the rim in (A). Despite markedly reduced MHC, inferior elongation of the cup is evident from left to right, with the reference plane retaining a similar level in the ONH (vascular trunk prominence represented similarly). Hence, the ONH and reference plane do not seem to have shifted in relation to each other. In (C), the reference plane is superficial to the vascular trunk (not seen) and the cup is wider. The reference plane in the right image appears to lie even more superficially than in the left image (to the vascular prominence) and has a wider cup. Rim appearance in (C) does not closely resemble that of (A). In (D, left), the z-axis position of the reference plane in relation to the vascular prominence is similar to that of (B). In the right image, however, the large reduction of MHC causes the reference plane to lie more superficially in the ONH (above the vessel trunk), rendering the cup diffusely larger and appearing different from (A, right).
Table 1.
 
Demographic Data for Normal and Glaucoma Groups
Table 1.
 
Demographic Data for Normal and Glaucoma Groups
Normal Glaucoma P
20 20
Age (y) 66.5 ± 9.2 68.8 ± 11.5 0.70
Male 12 10
Right eyes 13 10
Visual field MD (dB) +0.11 ± 1.2 −4.6 ± 1.9 0.0000
Visual field CPSD (dB) 0.89 ± 0.75 3.3 ± 1.8 0.0000
Mean SD (μm) of mean images 22.6 ± 8.1* 25.9 ± 9.65* 0.17*
22.9 ± 10.2 27.2 ± 10.5 0.16
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×