REVIEW URRENT C OPINION
Diagnosing glaucoma progression with optical coherence tomography Christopher Kai-Shun Leung
Purpose of review Optical coherence tomography (OCT) imaging of the retinal nerve fiber layer (RNFL), optic nerve head (ONH) and macula has gained popularity in recent years to detect and monitor glaucoma. With significant improvement in scan speed and scan resolution, spectral-domain OCT has become an efficient platform to evaluate progressive RNFL thinning and ONH remodeling. This review summarizes the recent progress of OCT RNFL, ONH and macular measurements for the detection of glaucoma progression. Recent findings The RNFL thickness map facilitates visualization of RNFL defects and their progression patterns, and attains a higher sensitivity to detect glaucoma progression compared with the conventional circumpapillary RNFL thickness profile. The measurement of lamina cribrosa displacement is informative to discern the relationship between intraocular pressure and ONH modeling. Ganglion cell and inner plexiform layer analysis has offered a useful alternative to track glaucoma progression. In the interpretation of progression analysis, the impact of age-related change, disease severity and image signal-to-noise ratio should always be considered. Summary The analyses of the RNFL thickness map, lamina cribrosa displacement and inner macular thickness have provided a new paradigm to evaluate glaucomatous damage and its progression. Keywords glaucoma progression, macular thickness, optic nerve head, optical coherence tomography, retinal nerve fiber layer
INTRODUCTION Optical coherence tomography (OCT) has rapidly evolved and become widely adopted for the detection of glaucomatous damage and progression since its introduction in 1991 [1]. Although the timedomain OCT is limited in scan speed (100–400 Ascans/s) and axial resolution (10 mm), most commercially available spectral-domain OCT (SD-OCT) instruments have a scan speed ranged between 26 000 and 53 000 A-scans/s and an axial resolution of approximately 5 mm (Table 1). The high scan speed and high image resolution is germane to minimizing test–retest measurement variabilities and facilitating the examination of the optic nerve head (ONH) and retina in high spatial resolution. These attributes are essential to detect glaucoma progression. In fact, it has been shown that the circumpapillary retinal nerve fiber layer (RNFL) thicknesses measured by the SD-OCT attained higher intervisit reproducibilities than the timedomain OCT measurements [2] and that the RNFL www.co-ophthalmology.com
thickness map with 200 200 pixel measurements over a 6 6 mm2 optic disc region is more sensitive than the circumpapillary RNFL thickness profile to detect glaucomatous RNFL defects [3]. Evaluation of progressive changes of the optic disc and RNFL should be based on event-analysis and/or trend-analysis [4]. In event analysis, progression is defined when the difference between the baseline and follow-up measurements of the parameter of interest is greater than its test–retest variability (or the reproducibility coefficient). In trend analysis, regression analysis is performed Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong, People’s Republic of China Correspondence to Christopher K.S. Leung, MD, MB, ChB, Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong, People’s Republic of China. Tel: +852 2762 3181; fax: +852 2715 9490; e-mail: [email protected] Curr Opin Ophthalmol 2014, 25:104–111 DOI:10.1097/ICU.0000000000000024 Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
OCT measurements for the detection of glaucoma progression and discusses their limitations in progression analysis.
KEY POINTS SD-OCT provides reproducible measurements of the RNFL, ONH and inner macula. The RNFL thickness change map can detect progressive RNFL thinning missed by serial analysis of circumpapillary RNFL measurements. Measurement of lamina cribrosa displacement facilitates the investigation of the relationship between intraocular pressure and ONH remodeling. SD-OCT measurement of inner macula thickness is a useful alternative to track glaucoma progression. Consideration of age-related change, disease severity and signal-to-noise ratio of OCT images is pivotal to analysis of glaucoma progression.
between the parameter of interest and time. Progression is commonly defined when a significant negative slope is detected using linear regression analysis. The slope of the regression line represents the rate of change of the parameter of interest, which is often useful to estimate the rate of disease progression. Although glaucoma progression can be analyzed with confocal scanning laser ophthalmoscopy measurement of optic disc surface topology [5] and scanning laser polarimetry measurement of RNFL retardance [6]. OCT is unique in providing measurements of the RNFL, ONH, as well as the inner macula for the assessment of glaucoma progression. This review highlights the recent advances of
DETECTION OF PROGRESSIVE RETINAL NERVE FIBER LAYER THINNING – THE CIRCUMPAPILLARY RETINAL NERVE FIBER LAYER THICKNESS Glaucoma is a chronic degenerative optic neuropathy characterized by progressive loss of retinal ganglion cells and remodeling of the ONH. Loss of retinal ganglion cell axons can be detected clinically as thinning of the RNFL, whereas remodeling of the ONH exhibits neuroretinal rim narrowing, optic disc excavation and displacement of the lamina cribrosa. Although OCT can capture both progressive RNFL thinning and ONH remodeling, the former is more widely adopted in current practice to monitor glaucoma progression. In fact, the first study reporting the use of OCT for glaucoma progression analysis examined the average circumpapillary RNFL thickness [7]. Using a prototype OCT device with an axial resolution of 10 mm and a scan speed of 40 A-scans/s, Wollstein et al. [7] measured the circumpapillary RNFL thickness (with a scan diameter of 3.4 mm) of 64 eyes of 37 glaucoma patients every 6 months for a median of 4.7 years. With progression defined using event analysis (a reduction of the average RNFL thickness of at least 20 mm from the baseline measurement in at least two consecutive visits), they detected progression in 22% of eyes by OCT, 9% by visual field and 3% by
Table 1. Commercially available spectral-domain optical coherence tomography systems RTVue SD-OCT
Cirrus HD-OCT
Spectralis OCT
Topcon 3D OCT
RS-3000 OCT
Envisu C-Class SDOIS
Manufacturer
Optovue
Carl Zeiss Meditec
Heidelberg Engineering
Topcon
NIDEK
Bioptigen
Superluminescent diode wavelength
840 10 nm
840 nm
870 nm
840 nm
880 nm
840 nm / 870 nm
Scan speed
26 000 A-scan/s
27 000 A-scan/s
40 000 A-scan/s
27 000 A-scan/s
53 000 A-scans/s
32 000 A-scans/s
Axial resolution
5 mm
5 mm
3.9 mm
5-6 mm
7 mm
4–6 mm
Transverse resolution
15 mm
15 mm
14 mm
20 mm
20 mm
20 mm
Progression reports
RNFL Change Analysis and GCC Progression Analysis
GPA of RNFL and ONH measurements
RNFL Change Report with FoDi
RNFL Trend Analysis
Multifunctional Follow-up allowing detection of changes in RNFL and [ NFLþGCLþIPL] thicknesses
Not available
Trend analysis / event analysis)
Yes / No
Yes / Yes
No / No
Yes / No
Yes / No
No / No
[NFLþGCLþIPL], a structure composed of nerve fiber layer, ganglion cell layer, and inner plexiform layer; FoDi, fovea-to-disc alignment; GCC, ganglion cell complex; GPA, guided progression analysis; ONH, optic nerve head; RNFL, retinal nerve fiber layer; SD-OCT, spectral-domain optical coherence tomography.
1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
105
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Glaucoma
both. Although the sensitivity and specificity of the OCT for the detection of glaucoma progression were unknown, this study provides the first account that demonstrates the feasibility of using OCT RNFL measurement to monitor glaucoma progression. Commercially available statistical packages for change analysis provide an important reference for clinicians to evaluate disease progression. The Guided Progression Analysis (GPA) was first introduced in the third generation of time-domain OCT (software version 5.0, Stratus OCT, Carl Zeiss Meditec) in 2008. The Stratus OCT GPA performs linear regression analysis between average circumpapillary RNFL thickness and time, and reports the rate of change and P-value in the analysis printout. Using the Stratus OCT GPA, we computed the rate of change of average RNFL thickness in 116 eyes of 64 glaucoma patients who had been followed for 3–5 years [8]. For eyes with a significant negative trend, the rate of change of average RNFL thickness ranged between 1.2 and 15.4 mm/year and the median loss was 3.3 mm/year. Following 253 eyes of 253 glaucoma patients and suspects for a mean of 4.0 years, Medeiros et al. estimated that the mean rates of change of average RNFL thickness measured by the Stratus OCT in eyes with and without progression evident in the visual field and/or optic disc stereophotographs were 0.72 mm/year and 0.04 mm/year, respectively [9]. The trend analysis is useful not only to determine if progression has occurred, but also the rate of progression. However, the Stratus OCT images analyzed in the GPA are not registered. Measurement errors secondary to scan circle misalignment are inevitable [10], which considerably weaken the performance of the Stratus OCT in change detection. The application of eye-tracking (Spectralis OCT, Heidelberg Engineering) and the derivation of circumpapillary RNFL thickness profile from the same location in serially registered RNFL thickness maps (Cirrus HD-OCT, Carl Zeiss Meditec) are examples of strategies implemented in SD-OCT to improve change detection. We compared the performance of the time-domain (Stratus OCT, Carl Zeiss Meditec) and SD-OCT (Cirrus HD-OCT, Carl Zeiss Meditec) to detect progressive reduction of the average RNFL thickness using trend analysis in 128 eyes of 81 glaucoma patients followed for at least 2 years, and showed that the SD-OCT detected significantly more progressing eyes (21 eyes) than the time-domain OCT (4 eyes) (P < 0.001) [11]. Remarkably, five eyes were noted to have a significant positive trend when the analysis was computed with time-domain OCT measurements, whereas none showed increasing RNFL measurements over time using SD-OCT measurements, supporting that 106
www.co-ophthalmology.com
the SD-OCT is more reliable to evaluate progressive RNFL thinning than the time-domain OCT. To date, most longitudinal studies investigating OCT RNFL measurements in glaucoma progression are based upon the 3.46 mm circle scan. Notably, the selection of the 3.46 mm scan circle diameter is largely arbitrary, which is related to a previous study reporting that the average RNFL thickness derived from the 3.46 mm scan circle was more reproducible than measurements derived from other circle sizes [12]. However, that RNFL defects can be missed by the circumpapillary RNFL thickness profile indicates the importance of examination of the RNFL in areas other than the scan circle for monitoring glaucoma progression (Fig. 1) [3]. The analysis of the RNFL thickness map is thus more useful to visualize the evolvement of RNFL defects in progressing glaucoma patients.
DETECTION OF PROGRESSIVE RETINAL NERVE FIBER LAYER THINNING – THE RETINAL NERVE FIBER LAYER THICKNESS MAP The high scan speed of the SD-OCT allows high density sampling of RNFL thickness in the optic disc region, facilitating the construction of the RNFL thickness map for examination of the distribution pattern of the RNFL. Currently, only one SD-OCT instrument (Cirrus HD-OCT, Carl Zeiss Meditec) affords change analysis of the RNFL thickness maps. The GPA of the Cirrus HD-OCT performs event analysis of the RNFL thickness maps and displays the area of change in the RNFL thickness change map in addition to reporting trend and event analyses of the circumpapillary RNFL thickness profiles. It computes the differences in RNFL thicknesses between the follow up and two baseline images (baseline measurement minus follow up measurement) for each of the individual 50 50 superpixels (1 superpixel ¼ 4 4 pixels) encompassed in the 6 6 mm2 RNFL thickness map. If the difference of a particular superpixel is greater than the test–retest variability, the superpixel would be encoded in yellow in the RNFL thickness change map. When the difference is confirmed in a consecutive follow-up visit, the superpixel would be encoded in red (Fig. 2). The analysis printout displays the color coding in the RNFL thickness change map when there are at least 20 contiguous pixels showing significant thickness change. With event analysis of serial RNFL thickness maps, it is feasible to discern different patterns of progressive RNFL thinning in glaucoma. In a longitudinal study following 186 eyes of 103 glaucoma patients every 4 months for a mean of 44 months, the most Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
(a) Optic disc photograph
(b) Visual field pattern deviation plot
(d) Cirrus HD-OCT clock hour and average RNFL thickness
(c) Stratus OCT clock hour and average RNFL thickness 161 111 106 106 104 75
157 93
86
108
103 89
65
103.26
87 79 126 106 92
(e) Cirrus HD-OCT RNFL thickness deviation map
76
81
94
74 106 89 83
Microns 300 200 100 0 0 40 80 120 160 200 240 20 60 100 140 180 220 SUP NAS INF TEMP TEMP
Microns 200 100 0
0
TEMP
30
60 SUP
90 120 150 180 210 240 NAS
INF
TEMP
FIGURE 1. (a) A glaucomatous eye with an inferotemporal retinal nerve fiber layer (RNFL) defect barely visible in the fundus photograph (a) had superonasal defects in the visual field pattern deviation plot (b). Although the Stratus OCT (Carl Zeiss Meditec, Dublin, CA, USA) (c) and the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA, USA) (d) clock hour and average RNFL thicknesses fail to show any abnormality (all were within normal limits), the inferotemporal RNFL defect is evident in the RNFL thickness deviation map with abnormal pixels of RNFL measurement encoded in yellow and red. The RNFL thickness map (e) is more informative to detect RNFL defects than the circumpapillary RNFL measurements. Adapted from Ref. [3].
common patterns of progressive RNFL thinning was the expansion of RNFL defects followed by the development of new defects and then deepening of pre-existing defects [13 ]. More importantly, this &&
(a)
study shows that the most common locations of progressive RNFL thinning reside at the inferotemporal meridians at approximately 2 mm, but not at 1.73 mm, from the optic disc center, affirming the
(b)
(c)
(d) 8/21/2007 11:00:00 AM 4000–1012
12/21/2007 12:19:29 PM 4000–1012
4/25/2008 9:55:22 AM 4000–2012
8/19/2008 9:48:26 AM 4000–1012
12/22/2008 10:34:44 AM 4000–1012
4/21/2009 10:28:17 AM 4000–1012
8/21/2009 10:40:14 AM 4000–1012
12/17/2009 10:12:58 AM 4000–1012
4/22/2010 11:17:22 AM 4000–1012
8/23/2010 11:29:22 AM 4000–1012
12/21/2010 11:31:00 AM 4000–1012
4/21/2011 11:40:04 AM 4000–1012
10/3/2011 10:56:39 AM 4000–1012
9/10
10/10 Registered
9/10 Registered
9/10 Registered
9/10 Registered
9/10 Registered
8/10 Registered
8/10 Registered
8/10 Registered
9/10 Registered
9/10 Registered
7/10 Registered
8/10 Registered
Baseline 1
Baseline 2
FIGURE 2. (a) Optic disc photograph, (b) optical coherence tomography retinal nerve fiber layer (RNFL) thickness map, (c) RNFL thickness deviation map and (d) guided progression analysis printout (GPA, Carl Zeiss Meditec, Dublin, CA, USA) of a glaucomatous eye with an inferotemporal RNFL defect. The RNFL thickness map (b) is a topology map (red signifies thick and blue signifies thin RNFL measurements) showing the RNFL distribution profile at the 6 6 mm2 optic disc region. The RNFL thickness deviation map (c) indicates the pixels with abnormal RNFL thickness. A pixel would be coded in yellow or red if the RNFL measurement is below the lower 95 and 99% of the centile ranges for that particular pixel, respectively. The GPA (Carl Zeiss Meditec, Dublin, CA, USA) printout shows serial RNFL thickness maps (upper panel) and RNFL thickness change maps (lower panel) of the same eye (d). Pixels with RNFL thickness difference exceeding the test–retest variability between a follow up and the first and the second baseline images would be coded in yellow. If the same changes are evident in a consecutive follow-up image, the pixels would be coded in red. Significant progressive loss of the RNFL is noted at the inferotemporal sector. Adapted from Ref. [13 ]. &&
1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
107
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Glaucoma
importance of examining the RNFL thickness map instead of the circumpapillary RNFL thickness profile to detect change.
OPTIC NERVE HEAD CHANGES IN GLAUCOMA PROGRESSION Objective assessment of ONH changes has been largely evaluated with confocal scanning laser ophthalmoscopy and only a few studies have investigated the longitudinal change of the neuroretinal rim and optic cup in glaucoma with OCT [9,14,15]. Using the Stratus OCT (Carl Zeiss Meditec), Medeiros et al. [9] followed 253 glaucoma patients and suspects for a median of 4.0 years and showed that the rate of change of cup area was different between progressors and nonprogressors (defined with reference to optic disc stereophotographs and visual fields). We compared the rates of change of global neuroretinal rim area and average circumpapillary RNFL thickness measured by the Stratus OCT (Carl Zeiss Meditec) in 108 eyes of 70 glaucoma patients followed for a median of 3.2 years and found that there were large variations in the rates of change of these parameters, and the agreement of progression detection between neuroretinal rim and RNFL measurements was poor [14]. Among 10 eyes (9.3%) with RNFL progression and 16 eyes (14.8%) with neuroretinal rim progression, only 1 (0.9%) had progression evident in both. Progressive optic disc changes may not be coherent with progressive RNFL thinning in glaucoma progression even in the same eyes. The improved resolution of optic disc images captured with SD-OCT permits visualization and detection of the Bruch’s membrane opening, providing new insights into the definitions of optic disc margin and the neuroretinal rim [16 ,17–19]. Detection of the lamina cribrosa has also become feasible with SD-OCT imaging (Fig. 3) [20–22,23 ]. Lee et al. [23 ] demonstrated a reduction in the lamina &
&
&
(a)
cribrosa depth, defined as the distance from the line joining the Bruch’s membrane openings to the anterior surface of the lamina cribrosa, in 100 eyes of 100 glaucoma patients after the initiation intraocular pressure lowering therapy (62 with medical therapy, 38 with trabeculectomy), and the degree of reduction was associated with age and intraocular pressure. Active clinical and experimental research is ongoing with SD-OCT imaging of the ONH and lamina cribrosa to understand the process of ONH remodeling, the stress and strain responses of the lamina cribrosa, and their relationships with progressive axonal degeneration of retinal ganglion cells in glaucoma progression [24–27].
MEASUREMENT OF MACULAR THICKNESSES FOR DETECTION OF GLAUCOMA PROGRESSION Having the highest density of retinal ganglion cells in the retina, the macula is a strategic location for monitoring glaucoma progression. Macular measurements have been investigated with both time-domain and SD-OCT instruments for the detection of glaucomatous damage, but studies evaluating its application for disease progression are sparse. Although time-domain OCT did not show a significant difference in the rate of change of average macular thickness (an average of six radial scan lines, each with 6 mm long) between progressors and nonprogressors (defined as eyes with and without evidence of progression in the visual field and/or optic disc stereophotographs, respectively) [9], a recent study performed with the SD-OCT showed otherwise. Using similar definitions of progressors and nonprogressors, Sung et al. [28] examined 98 glaucoma patients with advanced glaucoma (mean baseline visual field MD ¼ 14.3 5.5 dB) followed for a mean of 2.2 years and reported a significant difference in the rate of change of average macular thickness (an average of (b)
FIGURE 3. Cross-sectional image of the optic disc obtained with spectral-domain optical coherence tomography (Spectralis OCT, Heidelberg Engineering, GmbH, Dossenheim, Germany) (a). The anterior surface of the lamina cribrosa is highlighted in orange and the Bruch’s membrane openings (BMO) are marked with red dots (b). Lamina cribrosa displacement (orange arrows) can be evaluated by measuring the distance between the line joining the BMOs (red line) and the anterior surface of the lamina cribrosa. 108
www.co-ophthalmology.com
Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
128 scan lines covering a 6 6 mm2 area), but not in the rate of change of average circumpapillary RNFL thickness, between progressors and nonprogressors. However, in their subsequent study evaluating 162 glaucoma patients with mild glaucoma (visual field MD ¼ 4.30 dB in progressors, and 9.84 dB in nonprogressors) followed for the same duration, significant differences in the rates of change of both circumpapillary RNFL and macular thicknesses between progressors and nonprogressors were found [15]. These studies reveal the potential utility of macular thickness measurement and underscore the limitation of OCT RNFL measurements for the detection of progression in the late stages of glaucoma. Notably, the outer retinal layers may not be involved in glaucomatous damage. Measurements of the macular nerve fiber, ganglion cell and inner plexiform layer thicknesses have been shown to be useful to monitor glaucoma progression (Fig. 4) [29]
LIMITATION OF OPTICAL COHERENCE TOMOGRAPHY MEASUREMENTS TO FOLLOW PROGRESSION It is worth noting that not all progressive changes of the optic disc, RNFL and macular thicknesses detected by the OCT are disease-related (Fig. 5). Age-related RNFL and macular thinning has been demonstrated in prospective longitudinal studies [29,30 ]. It has been estimated that the mean rates of age-related decline of average RNFL thickness and inner macular thickness were 0.52 mm/year [95% confidence interval (CI), 0.86 to 0.17) and 0.245 mm/year (95% CI, 0.389 to 0.100 mm/ year), respectively, with adjustment for baseline measurement, spherical error, optic disc area (for &
2/16/2009 8:06:40 PM 4000–1012 SS: 8/10 Average thickness: 93
(a)
6/26/2009 10:17:31 AM 4000–1012 R2 SS: 10/10 Average thickness: 92
12/17/2009 6:31:02 PM 4000–1012 R2 SS: 9/10 Average thickness: 86
RNFL thickness) and signal strength. In other words, detecting a significant negative slope in trend analysis of average RNFL thickness or macular thickness may not necessarily signify glaucoma progression. In a prospective study following 150 eyes of 90 glaucoma patients every 4 months for a mean of 3.8 years, 50.0, 30.0 and 27.3% of eyes showed progression by trend analyses of the inner macular, total macular and circumpapillary RNFL thicknesses, respectively [29]. After accounting for agerelated changes, the proportions of eyes detected with progression decreased to 20.0% for the inner retinal thickness, 16.0% for the total macular thickness and 26.7% for the RNFL thickness. This finding underlines the impact of age-related change on macular and RNFL measurements. Taking agerelated decline into consideration is of particular importance in the change analysis of macular thicknesses in glaucoma patients. OCT is limited to detect progressive RNFL thinning in advanced optic neuropathies. This is supported by the findings that the rate of change of RNFL thickness is associated with the baseline measurement (a faster rate of reduction of RNFL thickness is found in eyes with a thicker RNFL) [29,30 ] and that the RNFL thickness does not fall below 30 mm even in eyes with end-stage optic neuropathies with no light perception [31]. That OCT RNFL segmentation could be less robust and the contribution of retinal blood vessels to the RNFL thickness measurement is more substantive when the RNFL is thin have rendered OCT less sensitive to measure progressive RNFL thinning in the late stages of glaucoma. OCT measurements are positively associated with the signal-to-noise ratio (or the signal strength)
6/25/2010 10:57:23 AM 4000–1012 R2 SS: 9/10 Average thickness: 87
&
4/15/2011 2:51:15 PM 4000–1012 R2 SS: 8/10 Average thickness: 86
12/12/2011 10:29:13 AM 4000–1012 R2 SS: 8/10 Average thickness: 83
4/16/2012 10:41:06 AM 4000–1012 R2 SS: 7/10 Average thickness: 79
8/9/2012 10:59:12 AM 4000–1012 R2 SS: 8/10 Average thickness: 74
350 175 0 µm
(b)
Baseline 1
Baseline 2
(c)
FIGURE 4. Progressive superior retinal nerve fiber layer (RNFL) thinning of a glaucomatous eye is detected in the RNFL thickness maps (a) and RNFL thickness change maps (b) obtained with a spectral-domain optical coherence tomography (Cirrus HD-OCT, Carl Zeiss Meditec, Dublin, CA, USA). Although the instrument software does not provide progression analysis of macular thicknesses, progressive reduction of ganglion cell inner plexiform layer (GCIPL) over the superior macula can be appreciated in the GCIPL thickness maps (c). 1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
109
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Glaucoma
7/7/2008 4:27:30 PM 4000–1012 8/10
11/11/2008 5:00:27 PM 4000–1012 8/10 Registered
Baseline 1
Baseline 2
7/9/2009 10:06:38 AM 4000–1012 9/10 Registered
11/13/2009 10:17:04 AM 4000–1012 9/10 Registered
3/18/2010 10:54:17 AM 4000–1012 8/10 Registered
7/15/2010 12:19:30 PM 4000–1012 8/10 Registered
11/18/2010 10:52:18 AM 4000–1012 8/10 Registered
3/17/2011 11:17:50 AM 4000–1012 9/10 Registered
350 175 0 µm
Average RNFL thickness
RNFL thickness profiles µm
µm 150 100 50 0
B_1
B_2
C
200 59
60
61
62
63
64 Age (years)
200
Overall thickness
0
µm 150 100 50 0
0 TEMP
59
60
61
62
63
30
60 90 SUP
120 150 NAS
180 210 INF
240 TEMP
64 Age (years)
Superior thickness Rate of change: –6.6 +/– 4.6 µm/year Statistically significant P < 0.05
RNFL summary OS RNFL thickness map progression
µm 150 100 50 0
RNFL thickness profiles progression 59
60
61
62
63
64 Age (years)
Inferior thickness
Average RNFL thickness progression Possible loss
Likely loss
Possible increase
FIGURE 5. Guided progression analysis (GPA, Carl Zeiss Meditec, Dublin, CA, USA) printout of a normal healthy eye followed from July 2008 to March 2011. Progressive superotemporal retinal nerve fiber layer (RNFL) thinning is detected in the RNFL thickness maps, the RNFL thickness change maps and the RNFL thickness profiles during the follow-up period. There is also a significant negative trend (rate of change ¼ 6.6 4.6 mm/year) in the analysis of superior RNFL thickness against age. However, optic disc assessment showed no signs of glaucomatous damage. The intraocular pressure was all along below 21 mmHg and visual field testing was normal. This eye shows evidence of progressive age-related RNFL thinning.
of the OCT images [32–34]. As the signal-to-noise ratio of OCT images may decrease over time with the development of media opacities (e.g. cataract, vitreous opacities), interpretation of glaucoma progression should always take the potential influence of the signal-to-noise ratio of the image series into consideration.
SUMMARY Among all the commercially available digital imaging technologies, SD-OCT is unique in being able to 110
www.co-ophthalmology.com
image both progressive degeneration of the optic nerve and remodeling of the ONH. The RNFL thickness map facilitates the analysis of RNFL defects and its progression patterns, improving the sensitivity to detect glaucoma compared with the conventional circumpapillary RNFL measurement. Visualization of the deep ONH structures, including the Bruch’s membrane opening and the lamina cribrosa, has changed the landscape in the investigation of ONH remodeling. It is conceivable that integrating change analyses of ONH, RNFL and macular Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
thicknesses would enrich our understanding of glaucoma progression and provide a new platform to monitor disease progression in glaucoma care. Acknowledgements None. Conflicts of interest CL Speaker honorarium Carl Zeiss Meditec, Heidelberg Engineering; Research support – Carl Zeiss Meditec, Optovue. There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science 1991; 254:1178–1181. 2. Leung CK, Cheung CY, Weinreb RN, et al. Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: a variability and diagnostic performance study. Ophthalmology 2009; 116:1257–1263. 3. Leung CK, Lam S, Weinreb RN, et al. Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: analysis of the retinal nerve fiber layer map for glaucoma detection. Ophthalmology 2010; 117:1684– 1691. 4. Weinreb RN, Garway-Heath D, Leung CK, Medeiros FA, Crowston JG, editors. Consensus Series 8: Progression. The Hague, The Netherlands: Kugler; 2011. 5. Chauhan BC, Blanchard JW, Hamilton DC, et al. Technique for detecting serial topographic changes in the optic disc and peripapillary retina using scanning laser tomography. Invest Ophthalmol Vis Sci 2000; 41:775– 782. 6. Medeiros FA, Alencar LM, Zangwill LM, et al. Detection of progressive retinal nerve fiber layer loss in glaucoma using scanning laser polarimetry with variable corneal compensation. Invest Ophthalmol Vis Sci 2009; 50:1675– 1681. 7. Wollstein G, Schuman JS, Price LL, et al. Optical coherence tomography longitudinal evaluation of retinal nerve fiber layer thickness in glaucoma. Arch Ophthalmol 2005; 123:464–470. 8. Leung CK, Cheung CY, Weinreb RN, et al. Evaluation of retinal nerve fiber layer progression in glaucoma: a study on optical coherence tomography guided progression analysis. Invest Ophthalmol Vis Sci 2010; 51:217– 222. 9. Medeiros FA, Zangwill LM, Alencar LM, et al. Detection of glaucoma progression with stratus OCT retinal nerve fiber layer, optic nerve head, and macular thickness measurements. Invest Ophthalmol Vis Sci 2009; 50:5741–5748. 10. Cheung CY, Yiu CK, Weinreb RN, et al. Effects of scan circle displacement in optical coherence tomography retinal nerve fibre layer thickness measurement: a RNFL modelling study. Eye (Lond) 2009; 23:1436–1441. 11. Leung CK, Chiu V, Weinreb RN, Liu S, et al. Evaluation of retinal nerve fiber layer progression in glaucoma: a comparison between spectral-domain and timedomain optical coherence tomography. Ophthalmology 2011; 118:1558– 1562. 12. Schuman JS, Pedut-Kloizman T, Hertzmark E, et al. Reproducibility of nerve fiber layer thickness measurements using optical coherence tomography. Ophthalmology 1996; 103:1889–1898. 13. Leung CK, Yu M, Weinreb RN, et al. Retinal nerve fiber layer imaging with && spectral-domain optical coherence tomography: patterns of retinal nerve fiber layer progression. Ophthalmology 2012; 119:1858–1866. This study provides the first account of using the SD-OCT RNFL thickness maps to visualize different patterns of RNFL defect progression and illustrates that progression can be missed using the conventional circumpapillary RNFL scan.
14. Leung CK, Liu S, Weinreb RN, Lai G, et al. Evaluation of retinal nerve fiber layer progression in glaucoma a prospective analysis with neuroretinal rim and visual field progression. Ophthalmology 2011; 118:1551–1557. 15. Na JH, Sung KR, Lee JR, et al. Detection of glaucomatous progression by spectral-domain optical coherence tomography. Ophthalmology 2013; 120:1388–1395. 16. Chauhan BC, Burgoyne CF. From clinical examination of the optic disc to & clinical assessment of the optic nerve head: a paradigm change. Am J Ophthalmol 2013; 156:218–227. This review highlights the importance of SD-OCT in defining the optic disc margin and measuring the neuroretinal rim width. 17. Chauhan BC, O’Leary N, Almobarak FA, et al. Enhanced detection of openangle glaucoma with an anatomically accurate optical coherence tomographyderived neuroretinal rim parameter. Ophthalmology 2013; 120:535–543. 18. Mwanza JC, Oakley JD, Budenz DL, et al. Ability of cirrus HD-OCT optic nerve head parameters to discriminate normal from glaucomatous eyes. Ophthalmology 2011; 118:241–248. 19. Yang B, Ye C, Yu M, et al. Optic disc imaging with spectral-domain optical coherence tomography: variability and agreement study with Heidelberg retinal tomograph. Ophthalmology 2012; 119:1852–1857. 20. Srinivasan VJ, Adler DC, Chen Y, et al. Ultrahigh-speed optical coherence tomography for three-dimensional and en face imaging of the retina and optic nerve head. Invest Ophthalmol Vis Sci 2008; 49:5103–5110. 21. Agoumi Y, Sharpe GP, Hutchison DM, et al. Laminar and prelaminar tissue displacement during intraocular pressure elevation in glaucoma patients and healthy controls. Ophthalmology 2011; 118:52–59. 22. Lee EJ, Kim TW, Weinreb RN, et al. Reversal of lamina cribrosa displacement after intraocular pressure reduction in open-angle glaucoma. Ophthalmology 2013; 120:553–559. 23. Kim TW, Kagemann L, Girard MJ, et al. Imaging of the lamina cribrosa in & glaucoma: perspectives of pathogenesis and clinical applications. Curr Eye Res 2013; 38:903–909. This review provides succinct updates of the applications, significance and limitations of OCT imaging of the lamina cribrosa. 24. Strouthidis NG, Fortune B, Yang H, et al. Longitudinal change detected by spectral domain optical coherence tomography in the optic nerve head and peripapillary retina in experimental glaucoma. Invest Ophthalmol Vis Sci 2011; 52:1206–1219. 25. Fortune B, Burgoyne CF, Cull GA, et al. Structural and functional abnormalities of retinal ganglion cells measured in vivo at the onset of optic nerve head surface change in experimental glaucoma. Invest Ophthalmol Vis Sci 2012; 53:3939–3950. 26. Burgoyne CF. A biomechanical paradigm for axonal insult within the optic nerve head in aging and glaucoma. Exp Eye Res 2011; 93:120–132. 27. Burgoyne CF, Downs JC, Bellezza AJ, et al. The optic nerve head as a biomechanical structure: a new paradigm for understanding the role of IOPrelated stress and strain in the pathophysiology of glaucomatous optic nerve head damage. Prog Retin Eye Res 2005; 24:39–73. 28. Sung KR, Sun JH, Na JH, et al. Progression detection capability of macular thickness in advanced glaucomatous eyes. Ophthalmology 2012; 119:308– 313. 29. Leung CK, Ye C, Weinreb RN, et al. Impact of age-related change of retinal nerve fiber layer and macular thicknesses on evaluation of glaucoma progression. Ophthalmology 2013; 120:2493–2500. 30. Leung CK, Yu M, Weinreb RN, et al. Retinal nerve fiber layer imaging with & spectral-domain optical coherence tomography: a prospective analysis of age-related loss. Ophthalmology 2012; 119:731–737. This paper describes the first prospective study reporting that progressive RNFL thinning can be detected in normal healthy eyes. Differentiation of age-related change form disease-related change is important in the analysis of glaucoma progression. 31. Chan CK, Miller NR. Peripapillary nerve fiber layer thickness measured by optical coherence tomography in patients with no light perception from longstanding nonglaucomatous optic neuropathies. J Neuroophthalmol 2007; 27:176–179. 32. Samarawickrama C, Pai A, Huynh SC, et al. Influence of OCT signal strength on macular, optic nerve head, and retinal nerve fiber layer parameters. Invest Ophthalmol Vis Sci 2010; 51:4471–4475. 33. Cheung CY, Leung CK, Lin D, et al. Relationship between retinal nerve fiber layer measurement and signal strength in optical coherence tomography. Ophthalmology 2008; 115:1347–1351. 34. Rao HL, Kumar AU, Babu JG, et al. Predictors of normal optic nerve head, retinal nerve fiber layer, and macular parameters measured by spectral domain optical coherence tomography. Invest Ophthalmol Vis Sci 2011; 52:1103– 1110.
1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
111
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing glaucoma progression with optical coherence tomography Christopher Kai-Shun Leung
Purpose of review Optical coherence tomography (OCT) imaging of the retinal nerve fiber layer (RNFL), optic nerve head (ONH) and macula has gained popularity in recent years to detect and monitor glaucoma. With significant improvement in scan speed and scan resolution, spectral-domain OCT has become an efficient platform to evaluate progressive RNFL thinning and ONH remodeling. This review summarizes the recent progress of OCT RNFL, ONH and macular measurements for the detection of glaucoma progression. Recent findings The RNFL thickness map facilitates visualization of RNFL defects and their progression patterns, and attains a higher sensitivity to detect glaucoma progression compared with the conventional circumpapillary RNFL thickness profile. The measurement of lamina cribrosa displacement is informative to discern the relationship between intraocular pressure and ONH modeling. Ganglion cell and inner plexiform layer analysis has offered a useful alternative to track glaucoma progression. In the interpretation of progression analysis, the impact of age-related change, disease severity and image signal-to-noise ratio should always be considered. Summary The analyses of the RNFL thickness map, lamina cribrosa displacement and inner macular thickness have provided a new paradigm to evaluate glaucomatous damage and its progression. Keywords glaucoma progression, macular thickness, optic nerve head, optical coherence tomography, retinal nerve fiber layer
INTRODUCTION Optical coherence tomography (OCT) has rapidly evolved and become widely adopted for the detection of glaucomatous damage and progression since its introduction in 1991 [1]. Although the timedomain OCT is limited in scan speed (100–400 Ascans/s) and axial resolution (10 mm), most commercially available spectral-domain OCT (SD-OCT) instruments have a scan speed ranged between 26 000 and 53 000 A-scans/s and an axial resolution of approximately 5 mm (Table 1). The high scan speed and high image resolution is germane to minimizing test–retest measurement variabilities and facilitating the examination of the optic nerve head (ONH) and retina in high spatial resolution. These attributes are essential to detect glaucoma progression. In fact, it has been shown that the circumpapillary retinal nerve fiber layer (RNFL) thicknesses measured by the SD-OCT attained higher intervisit reproducibilities than the timedomain OCT measurements [2] and that the RNFL www.co-ophthalmology.com
thickness map with 200 200 pixel measurements over a 6 6 mm2 optic disc region is more sensitive than the circumpapillary RNFL thickness profile to detect glaucomatous RNFL defects [3]. Evaluation of progressive changes of the optic disc and RNFL should be based on event-analysis and/or trend-analysis [4]. In event analysis, progression is defined when the difference between the baseline and follow-up measurements of the parameter of interest is greater than its test–retest variability (or the reproducibility coefficient). In trend analysis, regression analysis is performed Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong, People’s Republic of China Correspondence to Christopher K.S. Leung, MD, MB, ChB, Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong, People’s Republic of China. Tel: +852 2762 3181; fax: +852 2715 9490; e-mail: [email protected] Curr Opin Ophthalmol 2014, 25:104–111 DOI:10.1097/ICU.0000000000000024 Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
OCT measurements for the detection of glaucoma progression and discusses their limitations in progression analysis.
KEY POINTS SD-OCT provides reproducible measurements of the RNFL, ONH and inner macula. The RNFL thickness change map can detect progressive RNFL thinning missed by serial analysis of circumpapillary RNFL measurements. Measurement of lamina cribrosa displacement facilitates the investigation of the relationship between intraocular pressure and ONH remodeling. SD-OCT measurement of inner macula thickness is a useful alternative to track glaucoma progression. Consideration of age-related change, disease severity and signal-to-noise ratio of OCT images is pivotal to analysis of glaucoma progression.
between the parameter of interest and time. Progression is commonly defined when a significant negative slope is detected using linear regression analysis. The slope of the regression line represents the rate of change of the parameter of interest, which is often useful to estimate the rate of disease progression. Although glaucoma progression can be analyzed with confocal scanning laser ophthalmoscopy measurement of optic disc surface topology [5] and scanning laser polarimetry measurement of RNFL retardance [6]. OCT is unique in providing measurements of the RNFL, ONH, as well as the inner macula for the assessment of glaucoma progression. This review highlights the recent advances of
DETECTION OF PROGRESSIVE RETINAL NERVE FIBER LAYER THINNING – THE CIRCUMPAPILLARY RETINAL NERVE FIBER LAYER THICKNESS Glaucoma is a chronic degenerative optic neuropathy characterized by progressive loss of retinal ganglion cells and remodeling of the ONH. Loss of retinal ganglion cell axons can be detected clinically as thinning of the RNFL, whereas remodeling of the ONH exhibits neuroretinal rim narrowing, optic disc excavation and displacement of the lamina cribrosa. Although OCT can capture both progressive RNFL thinning and ONH remodeling, the former is more widely adopted in current practice to monitor glaucoma progression. In fact, the first study reporting the use of OCT for glaucoma progression analysis examined the average circumpapillary RNFL thickness [7]. Using a prototype OCT device with an axial resolution of 10 mm and a scan speed of 40 A-scans/s, Wollstein et al. [7] measured the circumpapillary RNFL thickness (with a scan diameter of 3.4 mm) of 64 eyes of 37 glaucoma patients every 6 months for a median of 4.7 years. With progression defined using event analysis (a reduction of the average RNFL thickness of at least 20 mm from the baseline measurement in at least two consecutive visits), they detected progression in 22% of eyes by OCT, 9% by visual field and 3% by
Table 1. Commercially available spectral-domain optical coherence tomography systems RTVue SD-OCT
Cirrus HD-OCT
Spectralis OCT
Topcon 3D OCT
RS-3000 OCT
Envisu C-Class SDOIS
Manufacturer
Optovue
Carl Zeiss Meditec
Heidelberg Engineering
Topcon
NIDEK
Bioptigen
Superluminescent diode wavelength
840 10 nm
840 nm
870 nm
840 nm
880 nm
840 nm / 870 nm
Scan speed
26 000 A-scan/s
27 000 A-scan/s
40 000 A-scan/s
27 000 A-scan/s
53 000 A-scans/s
32 000 A-scans/s
Axial resolution
5 mm
5 mm
3.9 mm
5-6 mm
7 mm
4–6 mm
Transverse resolution
15 mm
15 mm
14 mm
20 mm
20 mm
20 mm
Progression reports
RNFL Change Analysis and GCC Progression Analysis
GPA of RNFL and ONH measurements
RNFL Change Report with FoDi
RNFL Trend Analysis
Multifunctional Follow-up allowing detection of changes in RNFL and [ NFLþGCLþIPL] thicknesses
Not available
Trend analysis / event analysis)
Yes / No
Yes / Yes
No / No
Yes / No
Yes / No
No / No
[NFLþGCLþIPL], a structure composed of nerve fiber layer, ganglion cell layer, and inner plexiform layer; FoDi, fovea-to-disc alignment; GCC, ganglion cell complex; GPA, guided progression analysis; ONH, optic nerve head; RNFL, retinal nerve fiber layer; SD-OCT, spectral-domain optical coherence tomography.
1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
105
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Glaucoma
both. Although the sensitivity and specificity of the OCT for the detection of glaucoma progression were unknown, this study provides the first account that demonstrates the feasibility of using OCT RNFL measurement to monitor glaucoma progression. Commercially available statistical packages for change analysis provide an important reference for clinicians to evaluate disease progression. The Guided Progression Analysis (GPA) was first introduced in the third generation of time-domain OCT (software version 5.0, Stratus OCT, Carl Zeiss Meditec) in 2008. The Stratus OCT GPA performs linear regression analysis between average circumpapillary RNFL thickness and time, and reports the rate of change and P-value in the analysis printout. Using the Stratus OCT GPA, we computed the rate of change of average RNFL thickness in 116 eyes of 64 glaucoma patients who had been followed for 3–5 years [8]. For eyes with a significant negative trend, the rate of change of average RNFL thickness ranged between 1.2 and 15.4 mm/year and the median loss was 3.3 mm/year. Following 253 eyes of 253 glaucoma patients and suspects for a mean of 4.0 years, Medeiros et al. estimated that the mean rates of change of average RNFL thickness measured by the Stratus OCT in eyes with and without progression evident in the visual field and/or optic disc stereophotographs were 0.72 mm/year and 0.04 mm/year, respectively [9]. The trend analysis is useful not only to determine if progression has occurred, but also the rate of progression. However, the Stratus OCT images analyzed in the GPA are not registered. Measurement errors secondary to scan circle misalignment are inevitable [10], which considerably weaken the performance of the Stratus OCT in change detection. The application of eye-tracking (Spectralis OCT, Heidelberg Engineering) and the derivation of circumpapillary RNFL thickness profile from the same location in serially registered RNFL thickness maps (Cirrus HD-OCT, Carl Zeiss Meditec) are examples of strategies implemented in SD-OCT to improve change detection. We compared the performance of the time-domain (Stratus OCT, Carl Zeiss Meditec) and SD-OCT (Cirrus HD-OCT, Carl Zeiss Meditec) to detect progressive reduction of the average RNFL thickness using trend analysis in 128 eyes of 81 glaucoma patients followed for at least 2 years, and showed that the SD-OCT detected significantly more progressing eyes (21 eyes) than the time-domain OCT (4 eyes) (P < 0.001) [11]. Remarkably, five eyes were noted to have a significant positive trend when the analysis was computed with time-domain OCT measurements, whereas none showed increasing RNFL measurements over time using SD-OCT measurements, supporting that 106
www.co-ophthalmology.com
the SD-OCT is more reliable to evaluate progressive RNFL thinning than the time-domain OCT. To date, most longitudinal studies investigating OCT RNFL measurements in glaucoma progression are based upon the 3.46 mm circle scan. Notably, the selection of the 3.46 mm scan circle diameter is largely arbitrary, which is related to a previous study reporting that the average RNFL thickness derived from the 3.46 mm scan circle was more reproducible than measurements derived from other circle sizes [12]. However, that RNFL defects can be missed by the circumpapillary RNFL thickness profile indicates the importance of examination of the RNFL in areas other than the scan circle for monitoring glaucoma progression (Fig. 1) [3]. The analysis of the RNFL thickness map is thus more useful to visualize the evolvement of RNFL defects in progressing glaucoma patients.
DETECTION OF PROGRESSIVE RETINAL NERVE FIBER LAYER THINNING – THE RETINAL NERVE FIBER LAYER THICKNESS MAP The high scan speed of the SD-OCT allows high density sampling of RNFL thickness in the optic disc region, facilitating the construction of the RNFL thickness map for examination of the distribution pattern of the RNFL. Currently, only one SD-OCT instrument (Cirrus HD-OCT, Carl Zeiss Meditec) affords change analysis of the RNFL thickness maps. The GPA of the Cirrus HD-OCT performs event analysis of the RNFL thickness maps and displays the area of change in the RNFL thickness change map in addition to reporting trend and event analyses of the circumpapillary RNFL thickness profiles. It computes the differences in RNFL thicknesses between the follow up and two baseline images (baseline measurement minus follow up measurement) for each of the individual 50 50 superpixels (1 superpixel ¼ 4 4 pixels) encompassed in the 6 6 mm2 RNFL thickness map. If the difference of a particular superpixel is greater than the test–retest variability, the superpixel would be encoded in yellow in the RNFL thickness change map. When the difference is confirmed in a consecutive follow-up visit, the superpixel would be encoded in red (Fig. 2). The analysis printout displays the color coding in the RNFL thickness change map when there are at least 20 contiguous pixels showing significant thickness change. With event analysis of serial RNFL thickness maps, it is feasible to discern different patterns of progressive RNFL thinning in glaucoma. In a longitudinal study following 186 eyes of 103 glaucoma patients every 4 months for a mean of 44 months, the most Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
(a) Optic disc photograph
(b) Visual field pattern deviation plot
(d) Cirrus HD-OCT clock hour and average RNFL thickness
(c) Stratus OCT clock hour and average RNFL thickness 161 111 106 106 104 75
157 93
86
108
103 89
65
103.26
87 79 126 106 92
(e) Cirrus HD-OCT RNFL thickness deviation map
76
81
94
74 106 89 83
Microns 300 200 100 0 0 40 80 120 160 200 240 20 60 100 140 180 220 SUP NAS INF TEMP TEMP
Microns 200 100 0
0
TEMP
30
60 SUP
90 120 150 180 210 240 NAS
INF
TEMP
FIGURE 1. (a) A glaucomatous eye with an inferotemporal retinal nerve fiber layer (RNFL) defect barely visible in the fundus photograph (a) had superonasal defects in the visual field pattern deviation plot (b). Although the Stratus OCT (Carl Zeiss Meditec, Dublin, CA, USA) (c) and the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA, USA) (d) clock hour and average RNFL thicknesses fail to show any abnormality (all were within normal limits), the inferotemporal RNFL defect is evident in the RNFL thickness deviation map with abnormal pixels of RNFL measurement encoded in yellow and red. The RNFL thickness map (e) is more informative to detect RNFL defects than the circumpapillary RNFL measurements. Adapted from Ref. [3].
common patterns of progressive RNFL thinning was the expansion of RNFL defects followed by the development of new defects and then deepening of pre-existing defects [13 ]. More importantly, this &&
(a)
study shows that the most common locations of progressive RNFL thinning reside at the inferotemporal meridians at approximately 2 mm, but not at 1.73 mm, from the optic disc center, affirming the
(b)
(c)
(d) 8/21/2007 11:00:00 AM 4000–1012
12/21/2007 12:19:29 PM 4000–1012
4/25/2008 9:55:22 AM 4000–2012
8/19/2008 9:48:26 AM 4000–1012
12/22/2008 10:34:44 AM 4000–1012
4/21/2009 10:28:17 AM 4000–1012
8/21/2009 10:40:14 AM 4000–1012
12/17/2009 10:12:58 AM 4000–1012
4/22/2010 11:17:22 AM 4000–1012
8/23/2010 11:29:22 AM 4000–1012
12/21/2010 11:31:00 AM 4000–1012
4/21/2011 11:40:04 AM 4000–1012
10/3/2011 10:56:39 AM 4000–1012
9/10
10/10 Registered
9/10 Registered
9/10 Registered
9/10 Registered
9/10 Registered
8/10 Registered
8/10 Registered
8/10 Registered
9/10 Registered
9/10 Registered
7/10 Registered
8/10 Registered
Baseline 1
Baseline 2
FIGURE 2. (a) Optic disc photograph, (b) optical coherence tomography retinal nerve fiber layer (RNFL) thickness map, (c) RNFL thickness deviation map and (d) guided progression analysis printout (GPA, Carl Zeiss Meditec, Dublin, CA, USA) of a glaucomatous eye with an inferotemporal RNFL defect. The RNFL thickness map (b) is a topology map (red signifies thick and blue signifies thin RNFL measurements) showing the RNFL distribution profile at the 6 6 mm2 optic disc region. The RNFL thickness deviation map (c) indicates the pixels with abnormal RNFL thickness. A pixel would be coded in yellow or red if the RNFL measurement is below the lower 95 and 99% of the centile ranges for that particular pixel, respectively. The GPA (Carl Zeiss Meditec, Dublin, CA, USA) printout shows serial RNFL thickness maps (upper panel) and RNFL thickness change maps (lower panel) of the same eye (d). Pixels with RNFL thickness difference exceeding the test–retest variability between a follow up and the first and the second baseline images would be coded in yellow. If the same changes are evident in a consecutive follow-up image, the pixels would be coded in red. Significant progressive loss of the RNFL is noted at the inferotemporal sector. Adapted from Ref. [13 ]. &&
1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
107
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Glaucoma
importance of examining the RNFL thickness map instead of the circumpapillary RNFL thickness profile to detect change.
OPTIC NERVE HEAD CHANGES IN GLAUCOMA PROGRESSION Objective assessment of ONH changes has been largely evaluated with confocal scanning laser ophthalmoscopy and only a few studies have investigated the longitudinal change of the neuroretinal rim and optic cup in glaucoma with OCT [9,14,15]. Using the Stratus OCT (Carl Zeiss Meditec), Medeiros et al. [9] followed 253 glaucoma patients and suspects for a median of 4.0 years and showed that the rate of change of cup area was different between progressors and nonprogressors (defined with reference to optic disc stereophotographs and visual fields). We compared the rates of change of global neuroretinal rim area and average circumpapillary RNFL thickness measured by the Stratus OCT (Carl Zeiss Meditec) in 108 eyes of 70 glaucoma patients followed for a median of 3.2 years and found that there were large variations in the rates of change of these parameters, and the agreement of progression detection between neuroretinal rim and RNFL measurements was poor [14]. Among 10 eyes (9.3%) with RNFL progression and 16 eyes (14.8%) with neuroretinal rim progression, only 1 (0.9%) had progression evident in both. Progressive optic disc changes may not be coherent with progressive RNFL thinning in glaucoma progression even in the same eyes. The improved resolution of optic disc images captured with SD-OCT permits visualization and detection of the Bruch’s membrane opening, providing new insights into the definitions of optic disc margin and the neuroretinal rim [16 ,17–19]. Detection of the lamina cribrosa has also become feasible with SD-OCT imaging (Fig. 3) [20–22,23 ]. Lee et al. [23 ] demonstrated a reduction in the lamina &
&
&
(a)
cribrosa depth, defined as the distance from the line joining the Bruch’s membrane openings to the anterior surface of the lamina cribrosa, in 100 eyes of 100 glaucoma patients after the initiation intraocular pressure lowering therapy (62 with medical therapy, 38 with trabeculectomy), and the degree of reduction was associated with age and intraocular pressure. Active clinical and experimental research is ongoing with SD-OCT imaging of the ONH and lamina cribrosa to understand the process of ONH remodeling, the stress and strain responses of the lamina cribrosa, and their relationships with progressive axonal degeneration of retinal ganglion cells in glaucoma progression [24–27].
MEASUREMENT OF MACULAR THICKNESSES FOR DETECTION OF GLAUCOMA PROGRESSION Having the highest density of retinal ganglion cells in the retina, the macula is a strategic location for monitoring glaucoma progression. Macular measurements have been investigated with both time-domain and SD-OCT instruments for the detection of glaucomatous damage, but studies evaluating its application for disease progression are sparse. Although time-domain OCT did not show a significant difference in the rate of change of average macular thickness (an average of six radial scan lines, each with 6 mm long) between progressors and nonprogressors (defined as eyes with and without evidence of progression in the visual field and/or optic disc stereophotographs, respectively) [9], a recent study performed with the SD-OCT showed otherwise. Using similar definitions of progressors and nonprogressors, Sung et al. [28] examined 98 glaucoma patients with advanced glaucoma (mean baseline visual field MD ¼ 14.3 5.5 dB) followed for a mean of 2.2 years and reported a significant difference in the rate of change of average macular thickness (an average of (b)
FIGURE 3. Cross-sectional image of the optic disc obtained with spectral-domain optical coherence tomography (Spectralis OCT, Heidelberg Engineering, GmbH, Dossenheim, Germany) (a). The anterior surface of the lamina cribrosa is highlighted in orange and the Bruch’s membrane openings (BMO) are marked with red dots (b). Lamina cribrosa displacement (orange arrows) can be evaluated by measuring the distance between the line joining the BMOs (red line) and the anterior surface of the lamina cribrosa. 108
www.co-ophthalmology.com
Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
128 scan lines covering a 6 6 mm2 area), but not in the rate of change of average circumpapillary RNFL thickness, between progressors and nonprogressors. However, in their subsequent study evaluating 162 glaucoma patients with mild glaucoma (visual field MD ¼ 4.30 dB in progressors, and 9.84 dB in nonprogressors) followed for the same duration, significant differences in the rates of change of both circumpapillary RNFL and macular thicknesses between progressors and nonprogressors were found [15]. These studies reveal the potential utility of macular thickness measurement and underscore the limitation of OCT RNFL measurements for the detection of progression in the late stages of glaucoma. Notably, the outer retinal layers may not be involved in glaucomatous damage. Measurements of the macular nerve fiber, ganglion cell and inner plexiform layer thicknesses have been shown to be useful to monitor glaucoma progression (Fig. 4) [29]
LIMITATION OF OPTICAL COHERENCE TOMOGRAPHY MEASUREMENTS TO FOLLOW PROGRESSION It is worth noting that not all progressive changes of the optic disc, RNFL and macular thicknesses detected by the OCT are disease-related (Fig. 5). Age-related RNFL and macular thinning has been demonstrated in prospective longitudinal studies [29,30 ]. It has been estimated that the mean rates of age-related decline of average RNFL thickness and inner macular thickness were 0.52 mm/year [95% confidence interval (CI), 0.86 to 0.17) and 0.245 mm/year (95% CI, 0.389 to 0.100 mm/ year), respectively, with adjustment for baseline measurement, spherical error, optic disc area (for &
2/16/2009 8:06:40 PM 4000–1012 SS: 8/10 Average thickness: 93
(a)
6/26/2009 10:17:31 AM 4000–1012 R2 SS: 10/10 Average thickness: 92
12/17/2009 6:31:02 PM 4000–1012 R2 SS: 9/10 Average thickness: 86
RNFL thickness) and signal strength. In other words, detecting a significant negative slope in trend analysis of average RNFL thickness or macular thickness may not necessarily signify glaucoma progression. In a prospective study following 150 eyes of 90 glaucoma patients every 4 months for a mean of 3.8 years, 50.0, 30.0 and 27.3% of eyes showed progression by trend analyses of the inner macular, total macular and circumpapillary RNFL thicknesses, respectively [29]. After accounting for agerelated changes, the proportions of eyes detected with progression decreased to 20.0% for the inner retinal thickness, 16.0% for the total macular thickness and 26.7% for the RNFL thickness. This finding underlines the impact of age-related change on macular and RNFL measurements. Taking agerelated decline into consideration is of particular importance in the change analysis of macular thicknesses in glaucoma patients. OCT is limited to detect progressive RNFL thinning in advanced optic neuropathies. This is supported by the findings that the rate of change of RNFL thickness is associated with the baseline measurement (a faster rate of reduction of RNFL thickness is found in eyes with a thicker RNFL) [29,30 ] and that the RNFL thickness does not fall below 30 mm even in eyes with end-stage optic neuropathies with no light perception [31]. That OCT RNFL segmentation could be less robust and the contribution of retinal blood vessels to the RNFL thickness measurement is more substantive when the RNFL is thin have rendered OCT less sensitive to measure progressive RNFL thinning in the late stages of glaucoma. OCT measurements are positively associated with the signal-to-noise ratio (or the signal strength)
6/25/2010 10:57:23 AM 4000–1012 R2 SS: 9/10 Average thickness: 87
&
4/15/2011 2:51:15 PM 4000–1012 R2 SS: 8/10 Average thickness: 86
12/12/2011 10:29:13 AM 4000–1012 R2 SS: 8/10 Average thickness: 83
4/16/2012 10:41:06 AM 4000–1012 R2 SS: 7/10 Average thickness: 79
8/9/2012 10:59:12 AM 4000–1012 R2 SS: 8/10 Average thickness: 74
350 175 0 µm
(b)
Baseline 1
Baseline 2
(c)
FIGURE 4. Progressive superior retinal nerve fiber layer (RNFL) thinning of a glaucomatous eye is detected in the RNFL thickness maps (a) and RNFL thickness change maps (b) obtained with a spectral-domain optical coherence tomography (Cirrus HD-OCT, Carl Zeiss Meditec, Dublin, CA, USA). Although the instrument software does not provide progression analysis of macular thicknesses, progressive reduction of ganglion cell inner plexiform layer (GCIPL) over the superior macula can be appreciated in the GCIPL thickness maps (c). 1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
109
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Glaucoma
7/7/2008 4:27:30 PM 4000–1012 8/10
11/11/2008 5:00:27 PM 4000–1012 8/10 Registered
Baseline 1
Baseline 2
7/9/2009 10:06:38 AM 4000–1012 9/10 Registered
11/13/2009 10:17:04 AM 4000–1012 9/10 Registered
3/18/2010 10:54:17 AM 4000–1012 8/10 Registered
7/15/2010 12:19:30 PM 4000–1012 8/10 Registered
11/18/2010 10:52:18 AM 4000–1012 8/10 Registered
3/17/2011 11:17:50 AM 4000–1012 9/10 Registered
350 175 0 µm
Average RNFL thickness
RNFL thickness profiles µm
µm 150 100 50 0
B_1
B_2
C
200 59
60
61
62
63
64 Age (years)
200
Overall thickness
0
µm 150 100 50 0
0 TEMP
59
60
61
62
63
30
60 90 SUP
120 150 NAS
180 210 INF
240 TEMP
64 Age (years)
Superior thickness Rate of change: –6.6 +/– 4.6 µm/year Statistically significant P < 0.05
RNFL summary OS RNFL thickness map progression
µm 150 100 50 0
RNFL thickness profiles progression 59
60
61
62
63
64 Age (years)
Inferior thickness
Average RNFL thickness progression Possible loss
Likely loss
Possible increase
FIGURE 5. Guided progression analysis (GPA, Carl Zeiss Meditec, Dublin, CA, USA) printout of a normal healthy eye followed from July 2008 to March 2011. Progressive superotemporal retinal nerve fiber layer (RNFL) thinning is detected in the RNFL thickness maps, the RNFL thickness change maps and the RNFL thickness profiles during the follow-up period. There is also a significant negative trend (rate of change ¼ 6.6 4.6 mm/year) in the analysis of superior RNFL thickness against age. However, optic disc assessment showed no signs of glaucomatous damage. The intraocular pressure was all along below 21 mmHg and visual field testing was normal. This eye shows evidence of progressive age-related RNFL thinning.
of the OCT images [32–34]. As the signal-to-noise ratio of OCT images may decrease over time with the development of media opacities (e.g. cataract, vitreous opacities), interpretation of glaucoma progression should always take the potential influence of the signal-to-noise ratio of the image series into consideration.
SUMMARY Among all the commercially available digital imaging technologies, SD-OCT is unique in being able to 110
www.co-ophthalmology.com
image both progressive degeneration of the optic nerve and remodeling of the ONH. The RNFL thickness map facilitates the analysis of RNFL defects and its progression patterns, improving the sensitivity to detect glaucoma compared with the conventional circumpapillary RNFL measurement. Visualization of the deep ONH structures, including the Bruch’s membrane opening and the lamina cribrosa, has changed the landscape in the investigation of ONH remodeling. It is conceivable that integrating change analyses of ONH, RNFL and macular Volume 25 Number 2 March 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Diagnosing progression with optical coherence tomography Leung
thicknesses would enrich our understanding of glaucoma progression and provide a new platform to monitor disease progression in glaucoma care. Acknowledgements None. Conflicts of interest CL Speaker honorarium Carl Zeiss Meditec, Heidelberg Engineering; Research support – Carl Zeiss Meditec, Optovue. There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Huang D, Swanson EA, Lin CP, et al. Optical coherence tomography. Science 1991; 254:1178–1181. 2. Leung CK, Cheung CY, Weinreb RN, et al. Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: a variability and diagnostic performance study. Ophthalmology 2009; 116:1257–1263. 3. Leung CK, Lam S, Weinreb RN, et al. Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: analysis of the retinal nerve fiber layer map for glaucoma detection. Ophthalmology 2010; 117:1684– 1691. 4. Weinreb RN, Garway-Heath D, Leung CK, Medeiros FA, Crowston JG, editors. Consensus Series 8: Progression. The Hague, The Netherlands: Kugler; 2011. 5. Chauhan BC, Blanchard JW, Hamilton DC, et al. Technique for detecting serial topographic changes in the optic disc and peripapillary retina using scanning laser tomography. Invest Ophthalmol Vis Sci 2000; 41:775– 782. 6. Medeiros FA, Alencar LM, Zangwill LM, et al. Detection of progressive retinal nerve fiber layer loss in glaucoma using scanning laser polarimetry with variable corneal compensation. Invest Ophthalmol Vis Sci 2009; 50:1675– 1681. 7. Wollstein G, Schuman JS, Price LL, et al. Optical coherence tomography longitudinal evaluation of retinal nerve fiber layer thickness in glaucoma. Arch Ophthalmol 2005; 123:464–470. 8. Leung CK, Cheung CY, Weinreb RN, et al. Evaluation of retinal nerve fiber layer progression in glaucoma: a study on optical coherence tomography guided progression analysis. Invest Ophthalmol Vis Sci 2010; 51:217– 222. 9. Medeiros FA, Zangwill LM, Alencar LM, et al. Detection of glaucoma progression with stratus OCT retinal nerve fiber layer, optic nerve head, and macular thickness measurements. Invest Ophthalmol Vis Sci 2009; 50:5741–5748. 10. Cheung CY, Yiu CK, Weinreb RN, et al. Effects of scan circle displacement in optical coherence tomography retinal nerve fibre layer thickness measurement: a RNFL modelling study. Eye (Lond) 2009; 23:1436–1441. 11. Leung CK, Chiu V, Weinreb RN, Liu S, et al. Evaluation of retinal nerve fiber layer progression in glaucoma: a comparison between spectral-domain and timedomain optical coherence tomography. Ophthalmology 2011; 118:1558– 1562. 12. Schuman JS, Pedut-Kloizman T, Hertzmark E, et al. Reproducibility of nerve fiber layer thickness measurements using optical coherence tomography. Ophthalmology 1996; 103:1889–1898. 13. Leung CK, Yu M, Weinreb RN, et al. Retinal nerve fiber layer imaging with && spectral-domain optical coherence tomography: patterns of retinal nerve fiber layer progression. Ophthalmology 2012; 119:1858–1866. This study provides the first account of using the SD-OCT RNFL thickness maps to visualize different patterns of RNFL defect progression and illustrates that progression can be missed using the conventional circumpapillary RNFL scan.
14. Leung CK, Liu S, Weinreb RN, Lai G, et al. Evaluation of retinal nerve fiber layer progression in glaucoma a prospective analysis with neuroretinal rim and visual field progression. Ophthalmology 2011; 118:1551–1557. 15. Na JH, Sung KR, Lee JR, et al. Detection of glaucomatous progression by spectral-domain optical coherence tomography. Ophthalmology 2013; 120:1388–1395. 16. Chauhan BC, Burgoyne CF. From clinical examination of the optic disc to & clinical assessment of the optic nerve head: a paradigm change. Am J Ophthalmol 2013; 156:218–227. This review highlights the importance of SD-OCT in defining the optic disc margin and measuring the neuroretinal rim width. 17. Chauhan BC, O’Leary N, Almobarak FA, et al. Enhanced detection of openangle glaucoma with an anatomically accurate optical coherence tomographyderived neuroretinal rim parameter. Ophthalmology 2013; 120:535–543. 18. Mwanza JC, Oakley JD, Budenz DL, et al. Ability of cirrus HD-OCT optic nerve head parameters to discriminate normal from glaucomatous eyes. Ophthalmology 2011; 118:241–248. 19. Yang B, Ye C, Yu M, et al. Optic disc imaging with spectral-domain optical coherence tomography: variability and agreement study with Heidelberg retinal tomograph. Ophthalmology 2012; 119:1852–1857. 20. Srinivasan VJ, Adler DC, Chen Y, et al. Ultrahigh-speed optical coherence tomography for three-dimensional and en face imaging of the retina and optic nerve head. Invest Ophthalmol Vis Sci 2008; 49:5103–5110. 21. Agoumi Y, Sharpe GP, Hutchison DM, et al. Laminar and prelaminar tissue displacement during intraocular pressure elevation in glaucoma patients and healthy controls. Ophthalmology 2011; 118:52–59. 22. Lee EJ, Kim TW, Weinreb RN, et al. Reversal of lamina cribrosa displacement after intraocular pressure reduction in open-angle glaucoma. Ophthalmology 2013; 120:553–559. 23. Kim TW, Kagemann L, Girard MJ, et al. Imaging of the lamina cribrosa in & glaucoma: perspectives of pathogenesis and clinical applications. Curr Eye Res 2013; 38:903–909. This review provides succinct updates of the applications, significance and limitations of OCT imaging of the lamina cribrosa. 24. Strouthidis NG, Fortune B, Yang H, et al. Longitudinal change detected by spectral domain optical coherence tomography in the optic nerve head and peripapillary retina in experimental glaucoma. Invest Ophthalmol Vis Sci 2011; 52:1206–1219. 25. Fortune B, Burgoyne CF, Cull GA, et al. Structural and functional abnormalities of retinal ganglion cells measured in vivo at the onset of optic nerve head surface change in experimental glaucoma. Invest Ophthalmol Vis Sci 2012; 53:3939–3950. 26. Burgoyne CF. A biomechanical paradigm for axonal insult within the optic nerve head in aging and glaucoma. Exp Eye Res 2011; 93:120–132. 27. Burgoyne CF, Downs JC, Bellezza AJ, et al. The optic nerve head as a biomechanical structure: a new paradigm for understanding the role of IOPrelated stress and strain in the pathophysiology of glaucomatous optic nerve head damage. Prog Retin Eye Res 2005; 24:39–73. 28. Sung KR, Sun JH, Na JH, et al. Progression detection capability of macular thickness in advanced glaucomatous eyes. Ophthalmology 2012; 119:308– 313. 29. Leung CK, Ye C, Weinreb RN, et al. Impact of age-related change of retinal nerve fiber layer and macular thicknesses on evaluation of glaucoma progression. Ophthalmology 2013; 120:2493–2500. 30. Leung CK, Yu M, Weinreb RN, et al. Retinal nerve fiber layer imaging with & spectral-domain optical coherence tomography: a prospective analysis of age-related loss. Ophthalmology 2012; 119:731–737. This paper describes the first prospective study reporting that progressive RNFL thinning can be detected in normal healthy eyes. Differentiation of age-related change form disease-related change is important in the analysis of glaucoma progression. 31. Chan CK, Miller NR. Peripapillary nerve fiber layer thickness measured by optical coherence tomography in patients with no light perception from longstanding nonglaucomatous optic neuropathies. J Neuroophthalmol 2007; 27:176–179. 32. Samarawickrama C, Pai A, Huynh SC, et al. Influence of OCT signal strength on macular, optic nerve head, and retinal nerve fiber layer parameters. Invest Ophthalmol Vis Sci 2010; 51:4471–4475. 33. Cheung CY, Leung CK, Lin D, et al. Relationship between retinal nerve fiber layer measurement and signal strength in optical coherence tomography. Ophthalmology 2008; 115:1347–1351. 34. Rao HL, Kumar AU, Babu JG, et al. Predictors of normal optic nerve head, retinal nerve fiber layer, and macular parameters measured by spectral domain optical coherence tomography. Invest Ophthalmol Vis Sci 2011; 52:1103– 1110.
1040-8738 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-ophthalmology.com
111
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
![[Principles of glaucoma diagnostics with optical coherence tomography].](/assets/img/hold.png)
