To evaluate repeatability and positional independence of optic nerve head (ONH) and retinal nerve fiber layer (RNFL) thickness measurements in sitting and supine body positions using portable spectral-domain optical coherence tomography (iVue SD-OCT; Optovue Inc).
Evaluation of diagnostic technology.
Sixty eyes of 30 subjects (10 healthy younger adults aged 20–27 years, 10 healthy older adults aged 50–66 years, and 10 glaucoma patients aged 38–82 years) were included prospectively. For each participant, all measurements were taken in a single session. After 5 minutes in the supine position, 5 scans were obtained from both eyes. Following a 5-minute sitting adaptation, 5 scans were then obtained in the sitting position. The same instrument was used for all measurements. Repeatability and correlation between supine and sitting measurements of 4 ONH and 3 RNFL parameters were assessed using intraclass correlation coefficients (ICC), concordance correlation coefficients (ρ), and Bland-Altman plots.
Measurements were highly repeatable within individual eyes, both for ONH (ICC range, 73%–99%) and RNFL (ICC range, 72%–99%) parameters. The correlation between supine and sitting ONH measurements was strong and ranged from ρ = 97%–99% (younger healthy) to ρ = 98%–99% (older healthy) and ρ = 84%–99% (glaucoma). Bland-Altman plots indicated good agreement between sitting and supine readings of ONH and RNFL parameters.
Repeatability of measurements of ONH and RNFL is high and measurements between sitting and supine are highly correlated. The ability of the iVue SD-OCT to evaluate ONH and RNFL parameters is good to excellent in both body positions.
Primary open-angle glaucoma (POAG) is a chronic neurodegenerative disease characterized by progressive thinning of the retinal nerve fiber layer (RNFL) and of the neuroretinal rim within the optic nerve head (ONH). Assessment of the ONH and the RNFL plays an essential role in the diagnosis and follow-up of glaucomatous patients. However, the clinical examination of these structures is often difficult, qualitative, and subjective. Several imaging technologies, such as optical coherence tomography (OCT), allow for quantitative measurements of ONH topography and RNFL thickness in a noninvasive and objective manner.
OCT technology has changed considerably in recent years with the incorporation of spectral-domain (SD) imaging that offers significant advantages over the time-domain (TD) OCT techniques. Several SD-OCT instruments have been commercialized in recent years. The working principles are similar for all these devices and involve a superluminescent diode scan that collects 3-dimensional images of the peripapillary region from which information about the RNFL thickness can be obtained. Despite differences in acquisition time and resolution, these instruments have been shown to have similar diagnostic abilities to detect glaucoma.
Until very recently, ophthalmic applications of OCT technology could be performed exclusively in the sitting position, limiting its use in patients noncompliant because of physical disabilities. The newly introduced iVue (Optovue Inc, Fremont, California, USA) is a portable SD-OCT device that enables imaging in different body positions. OCT imaging in the supine body position may induce new methodological errors that can influence measurements. Intraocular pressure (IOP) changes in the magnitude of 7 mm Hg or greater, as a result of medical or surgical IOP lowering, have been shown to affect measurements of the RNFL and ONH using confocal scanning laser ophthalmoscopy (CLSO). This has been explained by the effect of IOP reduction causing a significant reversal of ONH cupping, interpreted as an increase in rim area by the CLSO. In contrast, posture-induced changes in IOP are generally in the order of 4 to 5 mm Hg in healthy subjects and glaucoma patients over a 24-hour period. Therefore, it is important to understand whether small to moderate postural IOP variations affect OCT readings of ONH and RNFL parameters dynamically when measured by sensitive instruments for assessing these changes over time. Currently, it is not known whether these IOP changes might affect OCT topographic measurements.
Before a new device can be widely accepted for use in clinical practice, its repeatability and diagnostic accuracy should be validated. Estimating instrument repeatability is essential for quantifying small changes detectable for identifying and monitoring glaucomatous progression. The current study seeks to evaluate the short-term repeatability and the correlation between supine and sitting measurements of ONH and RNFL parameters obtained with a portable OCT in healthy participants and glaucoma patients.
Three groups of 10 individuals were included in this cross-sectional, observational study. Younger healthy subjects (age, 22.9 ± 2.6 years) (mean ± standard deviation) and older healthy subjects (53.9 ± 5.0 years) were recruited from university employees and their families. A group of patients with POAG (61.3 ± 11.9 years) was recruited from a university-based glaucoma clinic at the Shiley Eye Center, University of California San Diego. Exclusion criteria included myopia ≤−5 diopters and presence of ocular disease other than glaucoma. Eyes were classified as glaucomatous if they had repeatable (2 consecutive) abnormal SAP test results on the 24-2 program of the Humphrey Visual Field Analyzer (Carl Zeiss Meditec, Inc, Dublin, California, USA). An abnormal SAP result was defined as having a pattern standard deviation outside the 95% confidence limits or a glaucoma hemifield test result outside the reference range, regardless of the appearance of the optic disc. Healthy subjects had a normal visual field test (using Statpac II, Swedish interactive thresholding algorithm 24-2, Zeiss-Humphrey Field Analyzer, Carl Zeiss Meditec, Inc), an IOP ≤21 mm Hg, and no clinical signs of glaucomatous optic neuropathy on dilated slit-lamp fundus examination.
Optical Coherence Tomography
Images of the optic nerve and adjacent retina were obtained using a portable spectral-domain OCT (SD-OCT) system (iVue SD-OCT; Optovue Inc, Fremont, California, USA). The iVue SD-OCT uses a superluminescent diode scan with a center wavelength of 840 ± 10 nm to provide high-resolution images. The manufacturer’s 3D Disk protocol was used to collect data. A 6 × 6-mm raster scan centered on the optic disc and composed of 101 B-scans each composed of 512 A-scans (acquisition time, 2.2 seconds) was used. The resulting scan provides a 3D image of the optic disc and surrounding area. The analysis points are derived from a 3.45-mm-diameter area. The iVue SD-OCT is delivered on a slit lamp–style base but can be mounted on multidirectional rolling floor stand (iStand; Optovue) for universal positioning of the device. The stand’s articulating arm is designed to facilitate scanning in the supine and other positions. All SD-OCT scans were obtained through undilated pupils. Quality assurance checks were done and images that had RNFL segmentation algorithm failures, had motion artifacts, were poorly focused, were not centered, or had a scan score index (SSI) of less than 40 were excluded.
Two experienced examiners scanned all participants. Measurements of both eyes of each study participant were taken in the supine followed by the sitting position. Participants were instructed to lie flat on a bed looking up for 5 minutes. For this position, the scan head was attached to the arm of the iStand, facing downward and placed directly over the subject’s eye. Five scans were taken of the right eye followed by the left eye. After having assumed the sitting position for 5 minutes, another set of 5 scans of each eye was taken in the same order using the iVue table and mount.
Outcome measures were repeatability within the same body position and correlation between both body positions for 4 ONH parameters (optic disc area, cup area, rim area, and rim volume) and 3 RNFL parameters (overall average RNFL thickness, inferior, and superior). All RNFL values were sampled from a 3.45-mm-diameter circle centered on the optic disc. The optic cup was defined automatically by the iVue software as the intersection points of the nerve head inner boundary and a parallel line that is 150 μm above the connecting line of the RPE tips. All other parameters were defined automatically by the iVue software.
Paired t test, 1-way analysis of variance (ANOVA) with Scheffé test for multiple comparisons, and χ 2 test were used to compare different measures between subject groups. Repeatability was defined as the variation of repeat measurements in the same body position. Intraclass correlation coefficients (ICC) were used to test repeatability of 5 measurements per eye in each body position. Concordance correlation coefficients (CCC) were calculated to test the strength of the relationship between body position and each of the ONH and RNFL parameters. CCCs are based on the differences between the observations made on the same subject in different positions and thus evaluate the agreement between 2 readings by measuring the variation from the 45 degree line through the origin. For this reason, CCC analysis was preferred to other reported statistical methods for evaluating correlation, such as intraclass correlation coefficients and linear regression. A correlation between 0.4 and 0.75 is considered fair to good, whereas higher correlations are qualified as excellent agreement. The agreement between measurements in the sitting and supine positions was further graphically evaluated using Bland-Altman plots and the difference in slopes was analyzed using Pitcairn’s test of difference in variance. Finally, Pearson correlation coefficients were used to determine the association between image quality and ONH and RNFL measurements.
Based on a significance level of 5%, a type II error rate (β) of 10%, our study had an 82% power to detect a difference of 0.3 mm 2 between disc area (for a mean area of 1.5 mm 2 and a mean standard deviation of 0.4 mm 2 ) and a 96% power to detect a difference of 10 μm between the RNFL measurements obtained in each position (for a mean RNFL thickness of 100 μm and a mean standard deviation of 10 μm). The 10 μm difference was based on the minimum significant difference required to exceed the intramachine variability. All tests were 2-tailed and a P value less than .05 was considered statistically significant. All statistical analyses were performed with commercially available software (Stata version 10; StataCorp, College Station, Texas, USA).
This study included 60 eyes (40 healthy, 20 glaucomatous) from 30 individuals (20 healthy, 10 glaucomatous). Mean age was 22.9 ± 2.6 years (mean ± standard deviation) for the younger healthy subjects, 53.9 ± 5.0 years for the older healthy subjects, and 61.3 ± 11.9 years for the glaucomatous patients. The older subjects had a similar age range to the glaucoma patients ( P = .09). Glaucomatous eyes had a mean IOP of 16.9 ± 3.8 mm Hg under ocular hypotensive treatment and a mean visual field mean deviation (MD) of −1.6 ± 2.0 dB and a mean pattern standard deviation (PSD) of 4.0 ± 3.7 dB at the time of iVue SD-OCT examination compared to 15.3 ± 2.3 mm Hg ( P = 0.09), 0.6 ± 1.1 dB ( P < .001), and 1.6 ± 0.4 dB ( P < .001), respectively, in the older group. Table 1 demonstrates demographic characteristics of the study population. All participants in the older healthy group were female and the difference to the other 2 groups was statistically significant ( P < .001). There was no difference in ancestry between the groups. Five RNFL scans of 5 different eyes had to be excluded from the analysis because of motion artifacts despite adequate quality scores and valid ONH readings.
|Younger Healthy (N = 10, 20 Eyes)||Older Healthy (N = 10, 20 Eyes)||Glaucoma (N = 10, 20 Eyes)|
|Age, average, y (SD) a||22.9 (2.6)||53.9 (5.0)||61.3 (11.9)|
|Sex (female), n (%) b||4 (40%)||10 (100%)||6 (60%)|
|Ancestry, n (%)|
|White||5 (50%)||6 (60%)||6 (60%)|
|Hispanic||5 (50%)||3 (30%)||3 (30%)|
|Asian||0||1 (10%)||1 (10%)|
Variation of Repeat Measurements in the Same Body Position
Tables 2 and 3 show the repeatability of iVue SD-OCT ONH and RNFL thickness measurements for each body position. The ICCs for ONH parameters ranged from 86% (rim volume) to 99% (cup area) for the younger healthy group, 77% (rim area) to 99% (cup area) for the older healthy group, and 73% (rim area) to 99% (cup volume) for the glaucoma group. The ICCs for RNFL thickness parameters ranged from 72% (inferior RNFL thickness) to 99% (superior RNFL thickness) for the younger healthy group, 72% (inferior RNFL thickness) to 97% (superior RNFL thickness) for the older healthy group, and 92% (inferior RNFL thickness) to 99% (overall RNFL thickness) for the glaucoma group.
|Group||Position||Disc Area (μm 2 )||Cup Area (μm 2 )||Rim Area (μm 2 )||Rim Volume (μm 3 )|
|OD||Supine||0.96 (0.92–0.99)||0.98 (0.97–0.99)||0.95 (0.92–0.99)||0.97 (0.95–0.99)|
|Sitting||0.97 (0.94–0.99)||0.98 (0.96–0.99)||0.93 (0.87–0.99)||0.88 (0.78–0.98)|
|OS||Supine||0.96 (0.95–0.99)||0.99 (0.99–1.0)||0.94 (0.89–0.99)||0.86 (0.74–0.98)|
|Sitting||0.98 (0.96–1.0)||0.99 (0.99–1.0)||0.95 (0.91–0.99)||0.96 (0.95–1.0)|
|OD||Supine||0.93 (0.88–0.99)||0.98 (0.97–1.0)||0.96 (0.91–1.0)||0.97 (0.94–0.99)|
|Sitting||0.97 (0.93–0.99)||0.98 (0.97–1.0)||0.95 (0.91–0.99)||0.97 (0.93–0.99)|
|OS||Supine||0.93 (0.86–0.99)||0.98 (0.96–0.99)||0.89 (0.79–0.98)||0.95 (0.91–0.99)|
|Sitting||0.90 (0.81–0.99)||0.99 (0.97–1.0)||0.77 (0.59–0.95)||0.94 (0.88–0.99)|
|OD||Supine||0.84 (0.69–0.98)||0.96 (0.92–1.0)||0.75 (0.55–0.96)||0.84 (0.70–0.98)|
|Sitting||0.93 (0.86–0.99)||0.98 (0.96–1.0)||0.73 (0.51–0.95)||0.81 (0.64–0.97)|
|OS||Supine||0.94 (0.88–0.99)||0.97 (0.93–1.0)||0.82 (0.66–0.98)||0.89 (0.79–0.99)|
|Sitting||0.98 (0.96–1.0)||0.97 (0.93–1.0)||0.82 (0.66–0.98)||0.90 (0.80–0.99)|
|Group||Position||Overall RNFL||Superior RNFL||Inferior RNFL|
|OD||Supine||0.93 (0.87–0.99)||0.99 (0.97–1.0)||0.72 (0.51–0.93)|
|Sitting||0.97 (0.95–0.99)||0.99 (0.97–0.10)||0.90 (0.81–0.99)|
|OS||Supine||0.97 (0.95–1.0)||0.97 (0.94–1.0)||0.91 (0.83–0.99)|
|Sitting||0.98 (0.97–1.0)||0.98 (0.96–1.0)||0.98 (0.96–1.0)|
|Supine||0.87 (0.76–0.98)||0.97 (0.93–0.99)||0.72 (0.51–0.92)|
|Sitting||0.93 (0.87–0.99)||0.97 (0.95–0.99)||0.86 (0.75–0.98)|
|Supine||0.96 (0.92–0.99)||0.95 (0.91–0.99)||0.92 (0.85–0.99)|
|Sitting||0.94 (0.86–0.99)||0.95 (0.90–0.99)||0.94 (0.88–0.99)|
|OD||Supine||0.99 (0.98–1.0)||0.98 (0.97–1.0)||0.98 (0.95–1.0)|
|Sitting||0.98 (0.96–0.99)||0.98 (0.97–1.0)||0.92 (0.84–0.99)|
|OS||Supine||0.98 (0.95–1.0)||0.98 (0.96–1.0)||0.98 (0.96–1.0)|
|Sitting||0.99 (0.98–1.0)||0.97 (0.94–1.0)||0.98 (0.96–1.0)|
Variation of Measurements Between the Supine and Sitting Position
Table 4 shows the different ONH parameters of each group for the 2 body positions. Statistical comparison of mean values and standard deviations among groups did not show significant posture-related differences for any of the studied parameters. Table 5 shows concordance correlation coefficients (CCC) and their 95% limits of agreement for ONH measurements in healthy subjects and glaucoma patients. Overall, CCC values were high, ranging from 98% to 99% for all parameters. Table 6 shows the different RNFL thickness parameters of each group for the 2 body positions. The correlation between RNFL thickness measurements for supine vs sitting positions was also very strong. Overall, CCC values varied from 96% (inferior RNFL thickness) to 99% (overall average RNFL thickness) ( Table 7 ).
|Group||Position||Disc Area (μm 2 )||Cup Area (μm 2 )||Rim Area (μm 2 )||Rim Volume (μm 3 )|
|OD||Supine||1.85 (0.40)||0.54 (0.46)||1.31 (0.34)||0.25 (0.15)|
|Sitting||1.87 (0.41)||0.54 (0.45)||1.34 (0.32)||0.27 (0.16)|
|OS||Supine||1.90 (0.50)||0.53 (0.60)||1.38 (0.34)||0.29 (0.15)|
|Sitting||1.89 (0.51)||0.53 (0.61)||1.36 (0.35)||0.29 (0.15)|
|OD||Supine||1.98 (0.35)||0.73 (0.44)||1.25 (0.45)||0.20 (0.12)|
|Sitting||1.97 (0.36)||0.74 (0.47)||1.24 (0.42)||0.20 (0.12)|
|OS||Supine||1.91 (0.40)||0.73 (0.46)||1.18 (0.27)||0.19 (0.09)|
|Sitting||1.92 (0.36)||0.71 (0.46)||1.21 (0.22)||0.20 (0.10)|
|OD||Supine||1.61 (0.86)||1.07 (0.58)||0.55 (0.29)||0.05 (0.03)|
|Sitting||2.10 (0.42)||1.41 (0.51)||0.69 (0.15)||0.06 (0.03)|
|OS||Supine||1.95 (0.40)||1.14 (0.49)||0.81 (0.18)||0.08 (0.04)|
|Sitting||1.98 (0.40)||1.11 (0.50)||0.87 (0.18)||0.09 (0.03)|
|Group||Disc Area||Cup Area||Rim Area||Rim Volume|
|Younger healthy||0.97 (0.94–0.99)||0.99 (0.99–1.00)||0.98 (0.97–0.99)||0.98 (0.96–0.99)|
|Older healthy||0.98 (0.96–0.99)||0.99 (0.98–0.99)||0.98 (0.97–0.99)||0.99 (0.98–0.99)|
|Glaucoma||0.96 (0.94–0.99)||0.98 (0.96–0.99)||0.84 (0.71–0.97)||0.92 (0.85–0.99)|
|Overall||0.98 (0.97–0.99)||0.99 (0.99–1.00)||0.98 (0.96–0.99)||0.99 (0.98–0.99)|
|Group||Position||Overall RNFL (μm)||Superior RNFL (μm)||Inferior RNFL (μm)|
|OD||Supine||103.9 (8.3)||128.1 (16.5)||137.8 (15.5)|
|Sitting||104.3 (9.3)||127.7 (15.7)||141.3 (15.6)|
|OS||Supine||101.7 (8.7)||124.5 (14.4)||138.1 (12.1)|
|Sitting||101.6 (9.2)||126.2 (16.1)||139.1 (13.7)|
|OD||Supine||100.7 (3.9)||115.4 (12.8)||129.4 (9.0)|
|Sitting||102.0 (4.5)||116.5 (14.0)||131.8 (9.2)|
|OS||Supine||98.7 (5.3)||119.0 (12.2)||127.9 (10.2)|
|Sitting||98.3 (4.2)||117.5 (9.1)||129.5 (10.4)|
|OD||Supine||79.5 (13.4)||89.1 (18.0)||101.3 (20.6)|
|Sitting||78.7 (14.2)||90.5 (18.9)||99.7 (22.0)|
|OS||Supine||82.7 (12.6)||99.3 (15.4)||107.0 (19.2)|
|Sitting||84.1 (12.8)||99.3 (15.2)||108.8 (20.4)|