Optical coherence tomography (OCT) is a noninvasive imaging technology that provides cross-sectional images of the retina with high resolution. Quantitative measurements of the macula are obtained which are of clinical importance in monitoring the retinal condition objectively. There are many factors that affect these quantitative measurements like scan centration and sampling density. Apart from these, errors due to differences in axial length (AXL) causing ocular magnification effects has been documented. The transverse mirror in OCT is calibrated for an AXL of 24.46 mm. Inter-individual differences in AXL from 24.46 mm would result in magnification errors in the measurements made on OCT.
Effect of ocular magnification was first described when measuring the size of a retinal lesion or optic disc on fundus photographs. Later, it was found to affect the retinal nerve fiber layer (RNFL) analysis in OCT as the thickness is measured at a constant distance (1.7 mm) from optic disc center and this distance was found to vary with different AXL. Its impact on the measurements of macula was recently reported, yet it has not been applied in clinical practice when measuring the lateral dimensions of macular holes, choroidal neovascular membranes (CNVM) or geographic atrophy and when measuring choroidal thickness at various eccentricities from fovea. Absence of an effect of magnification related to AXL on RNFL and macular thickness has also been reported. Hence, we aimed at studying the effect of ocular magnification on macular measurements (both lateral and axial) made on OCT.
Different methods of correcting for ocular magnification using AXL, ocular refraction, keratometry have been described. Amidst these methods, we have used the AXL method which is most reliable.
Materials and Methods
One hundred and fifty-one eyes of 151 subjects from the normative study data on the foveal morphology conducted at our hospital during the period of June 2012 and March 2013 were included in the study. The study was approved by institutional review board and was in accordance with Declaration of Helsinki. The study population included the subjects who came for routine eye examination and volunteers. A written informed consent was obtained from all subjects before enrolling into the project.
All subjects underwent comprehensive eye examination and macular scanning using Cirrus high-definition OCT (Model 4000, Version 6, Carl Zeiss Meditec, Inc., Dublin, CA) and AXL measurement using Ocuscan RxP (Alcon Laboratories, Inc., CA). All scans were performed by the same observer and care was taken to avoid obtaining decentered or tilted images. Subjects aged between 18 and 80 years having best corrected visual acuity of 20/30 or above, refractive error less than ± 3.00 DS, cylindrical error less than −2.00 DC were included in the study. Subjects with any retinal or optic nerve pathology and history of retinal surgery were excluded. OCT images with signal strength <6 out of 10 units and artifacts were excluded.
Optical coherence tomography
Scan protocol and ocular magnification correction
Macular cube 512 × 128 scan protocol was used for scanning the macula. It consists of 6 mm × 6 mm square grid consisting of 128 B-scans with each B-scan having 512 A-scans. The spacing between 2 A-scans is 11.72 µm for an AXL of 24.46 mm which is the instrument's default setting. Due to inter-individual differences in AXL, the scan length and A-scan spacing would differ between subjects. In order to correct for this resultant ocular magnification, a correction factor was applied for obtaining the actual scan length. The formula used was t = p × q × s; where “t” is the true scan length, “p” is the magnification related to the instrument, “q” is the magnification related to eye and “s” is the default scan length. The value of P is 3.382 for the instrument calibrated AXL of 24.46 mm which is a constant for a telecentric imaging system. Ocular magnification factor “q” is calculated by the formula 0.01306 × (AXL-1.82) where 1.82 is the distance between corneal apex and 2nd principal point of the eye.
Automated macular measurements
A custom written program was used for extracting foveal parameters like foveal diameter, foveal slope which were the lateral measurements and foveal depth, an axial measurement [Fig. 1]. Raw data of retinal thickness at all A-scan positions of the cube scan were imported into Matlab. In Matlab, the retinal thickness data from the horizontal B-scan representing foveal center were taken. Foveal center was identified as the deepest point in the foveal depression and the central light reflex. The A-scan numbers on the foveal B-scan were converted to microns using the AXL corrected scan length. The B-scan profile was fitted with polynomials with root mean square error of <1. The maximum degree of the polynomial that was fitted to a sample was 9. Nasal and temporal foveal rims where the retinal thickness is maximum were located automatically, and the distance between the rims was taken as foveal diameter. Foveal depth was the distance between the foveal center and the line connecting the rims. For foveal slope, the first derivative of the polynomials was found from which the maximum foveal slope was taken. This was converted to degrees and the average of nasal and temporal slope was taken as a horizontal foveal slope [Fig. 2]. The foveal parameters uncorrected and corrected for ocular magnification were obtained from custom written algorithm in Matlab.
Statistical analysis was performed in MS Excel 2003 and SPSS (SPSS Inc, Version 14, Chicago). All parameters followed the normal distribution, tested using Kolmogorov–Smirnov test. Mean and standard deviation is given for all parameters. Paired t-test was used to compare foveal parameters pre- and post-ocular magnification correction. Correlation between AXL difference and magnitude of error in foveal parameters was assessed. Alpha error was kept at 5%.
Of 151 subjects included in the study, 84 (55.6%) were males. Mean age of the subjects was 43.9 ± 14.4 years. Spherical equivalent of refractive error ranged from −3.00D to +3.75D with 58 (38.4%) subjects being emmetropic. A total of 25 subjects (16.6%) were mild myopes and the remaining 68 subjects (45%) were hyperopes. Mean AXL was 22.95 ± 0.78 mm ranging from 21.35 to 24.98 mm. The scan length of OCT (6 mm) was corrected for ocular magnification and the calculated actual scan length ranged from 5.18 to 6.14 mm with a mean of 5.60 mm. The mean ocular magnification (q) was 0.276 ± 0.01 and mean total magnification (p × q) was 0.934 ± 0.03.
The effect of ocular magnification on OCT measurements was assessed on the axial and lateral measurements of fovea namely foveal depth and foveal diameter and slope respectively. Table 1 gives the comparison of ocular magnification corrected and uncorrected measurements of foveal parameters. The mean difference between corrected and uncorrected parameters was statistically significant for lateral measurements namely foveal diameter and foveal slope (P < 0.001). Foveal depth was not statistically different before and after ocular magnification correction (P = 0.178).
The larger the deviation of the subject's AXL from the default AXL, larger was the error in lateral measurements. AXL lesser than OCT calibrated AXL (24.46 mm) resulted in an overestimation of foveal diameter (Pearson correlation; r = 0.961, P < 0.001) and an underestimation of foveal slope measurement (Pearson correlation; r = −0.863, P < 0.001) when compared with corrected measurements [Figs. 3 and 4]. Difference in foveal depth between corrected and uncorrected measurement did not have any relation with AXL [Fig. 5]. An error of around 118 µm occurred in foveal diameter for every 1 mm difference in AXL. In our sample, the maximum error in foveal diameter was 390 µm for an AXL of 21.58 mm. An error of 0.6° occurred in foveal slope for every 1 mm difference in AXL. There was no significant error in foveal depth with difference in AXL.
This study assessed the effect of ocular magnification due to inter-individual differences in AXL on macular measurements made on OCT. Variations in AXL cause alterations in the OCT B-scan length and consequently, the measurements made on it due to ocular magnification. Unlike axial measurements which are made vertically in depth, lateral measurements that are made along the horizontal extent were affected by differences in AXL, which is in agreement with previous studies. Wang et al. have reported no effect of magnification related to AXL on RNFL and macular thickness, but the authors attribute it to the narrow distribution of AXL in their sample.
When lateral measurements were not corrected for differences in AXL, an individual with a shorter AXL was found to have an increased foveal diameter, and a decreased the foveal slope compared to actual dimensions. For instance, the actual OCT scan length of an eye with shorter AXL (say 22.43 mm) would be 5.46 mm instead of default 6 mm. Therefore, 512 A-scans which makes up one B-scan would be accommodated within this smaller scan length that result in smaller separation between 2 A-scans (10.66 µm for scan length of 5.46 mm instead of 11.72 µm for the default 6 mm scan). This results in a smaller foveal diameter after correcting for ocular magnification factor. Consequently, for such a small change in the A-scan, the change in the retinal thickness would be greater, thus, increasing the foveal slope value. After correcting for ocular magnification, lateral measurements were found to be larger for longer eyes and vice versa. This finding was consistent with studies on RNFL thickness that measured a larger scan radius on myopic eyes with longer AXL.
In our sample, most of the subjects had AXL lesser than the OCT calibrated AXL of 24.46 mm with only 6 (4%) subjects having a longer AXL. The degree of error due to ocular magnification depends on the distribution of AXL in the study population. As the subjects’ AXL deviated more from 24.46 mm, the error in foveal diameter and foveal slope increased. Our finding is in agreement with that of Wagner-Schuman et al. Reported effects of AXL on macular and RNFL thickness cannot be interpreted correctly without correcting for magnification factors.
Lateral dimensions measured on OCT such as size of a macular hole, extent of a CNVM, are of clinical importance in assessing prognosis of surgery and monitoring the disease. Retinal thickness displayed in Early Treatment Diabetic Retinopathy Study (ETDRS) retinal thickness map is also prone to changes due to differences in AXL because it consists of measurements taken at specific points measured laterally from the center, the concentric rings of diameter 1 mm, 3 mm and 6 mm. An overall error of 44.9 µm in nine ETDRS segments and a maximum error of 7.86 µm in central subfield have been reported.
Several methods have been proposed to correct for ocular magnification by incorporating the spectacle refraction, keratometry, or AXL, each method yielding different values. We used the AXL method to correct for ocular magnification as described by Bennett et al. who introduced modifications in Littmann's formula. Although correcting ocular magnification using AXL is reliable, keratometry value or refractive error can be used to get an approximate of the corrected measurements. There can be a combined effect of AXL and keratometry values on OCT measurements which have not been evaluated in this study. Kuo et al. observed that the curvature of the retina seen in OCT is not the true curvature of the eye and proposed a correction technique for ocular shape which also corrects for ocular magnification. Since our measurements were performed in the central macula, this smaller field may not be affected by ocular shape.
We assessed the effect of ocular magnification on measurements made on OCT and found that the lateral measurements tend to be erroneous for AXL different from the OCT calibrated value of 24.46 mm. Therefore, to estimate the actual dimension of a retinal lesion using OCT, especially lateral dimension, the measurement should be corrected for ocular magnification factor.
Dr. G Venkiteswaran – for providing ideas on image analysis in Matlab.
1. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, et al Optical coherence tomography
2. Larsson J, Zhu M, Sutter F, Gillies MC. Relation between reduction of foveal thickness and visual acuity in diabetic macular edema treated with intravitreal triamcinolone Am J Ophthalmol. 2005;139:802–6
3. Moussa K, Lee JY, Stinnett SS, Jaffe GJ. Spectral domain optical coherence tomography
-determined morphologic predictors of age-related macular degeneration-associated geographic atrophy progression Retina. 2013;33:1590–9
4. Lujan BJ, Rosenfeld PJ, Gregori G, Wang F, Knighton RW, Feuer WJ, et al Spectral domain optical coherence tomographic imaging of geographic atrophy Ophthalmic Surg Lasers Imaging. 2009;40:96–101
5. Odell D, Dubis AM, Lever JF, Stepien KE, Carroll J. Assessing errors inherent in OCT-derived macular thickness maps J Ophthalmol. 2011;2011:692574
6. Wagner-Schuman M, Dubis AM, Nordgren RN, Lei Y, Odell D, Chiao H, et al Race- and sex-related differences in retinal thickness and foveal pit morphology Invest Ophthalmol Vis Sci. 2011;52:625–34
7. Wang XY, Huynh SC, Burlutsky G, Ip J, Stapleton F, Mitchell P. Reproducibility of and effect of magnification on optical coherence tomography
measurements in children Am J Ophthalmol. 2007;143:484–8
8. Bennett AG, Rudnicka AR, Edgar DF. Improvements on Littmann's method of determining the size of retinal features by fundus photography Graefes Arch Clin Exp Ophthalmol. 1994;232:361–7
9. Sanchez-Cano A, Baraibar B, Pablo LE, Honrubia FM. Magnification characteristics of the Optical Coherence Tomograph STRATUS OCT 3000 Ophthalmic Physiol Opt. 2008;28:21–8
10. Nowroozizadeh S, Cirineo N, Amini N, Knipping S, Chang T, Chou T, et al Influence of correction of ocular magnification on spectral-domain OCT retinal nerve fiber layer measurement variability and performance Invest Ophthalmol Vis Sci. 2014;55:3439–46
11. Garway-Heath DF, Rudnicka AR, Lowe T, Foster PJ, Fitzke FW, Hitchings RA. Measurement of optic disc size: Equivalence of methods to correct for ocular magnification Br J Ophthalmol. 1998;82:643–9
12. Bengtsson B, Krakau CE. Correction of optic disc measurements on fundus photographs Graefes Arch Clin Exp Ophthalmol. 1992;230:24–8
13. Bayraktar S, Bayraktar Z, Yilmaz OF. Influence of scan radius correction for ocular magnification and relationship between scan radius with retinal nerve fiber layer thickness measured by optical coherence tomography
J Glaucoma. 2001;10:163–9
14. Kang SH, Hong SW, Im SK, Lee SH, Ahn MD. Effect of myopia on the thickness of the retinal nerve fiber layer measured by Cirrus HD optical coherence tomography
Invest Ophthalmol Vis Sci. 2010;51:4075–83
15. Leung CK, Cheng AC, Chong KK, Leung KS, Mohamed S, Lau CS, et al Optic disc measurements in myopia with optical coherence tomography
and confocal scanning laser ophthalmoscopy Invest Ophthalmol Vis Sci. 2007;48:3178–83
16. Huang D, Chopra V, Lu AT, Tan O, Francis B, Varma R, et al Does optic nerve head size variation affect circumpapillary retinal nerve fiber layer thickness measurement by optical coherence tomography
? Invest Ophthalmol Vis Sci. 2012;53:4990–7
17. Oner V, Aykut V, Tas M, Alakus MF, Iscan Y. Effect of refractive status on peripapillary retinal nerve fibre layer thickness: A study by RTVue spectral domain optical coherence tomography
Br J Ophthalmol. 2013;97:75–9
18. Mwanza JC, Durbin MK, Budenz DL, Girkin CA, Leung CK, Liebmann JM, et al Profile and predictors of normal ganglion cell-inner plexiform layer thickness measured with frequency-domain optical coherence tomography
Invest Ophthalmol Vis Sci. 2011;52:7872–9
19. Manjunath V, Taha M, Fujimoto JG, Duker JS. Choroidal thickness in normal eyes measured using Cirrus HD optical coherence tomography
Am J Ophthalmol. 2010;150:325–9.e1
20. Dubis AM, McAllister JT, Carroll J. Reconstructing foveal pit morphology from optical coherence tomography
imaging Br J Ophthalmol. 2009;93:1223–7
21. Salchow DJ, Hwang AM, Li FY, Dziura J. Effect of contact lens power on optical coherence tomography
of the retinal nerve fiber layer Invest Ophthalmol Vis Sci. 2011;52:1650–4
22. Huynh SC, Wang XY, Rochtchina E, Mitchell P. Distribution of macular thickness by optical coherence tomography
: Findings from a population-based study of 6-year-old children Invest Ophthalmol Vis Sci. 2006;47:2351–7
23. Ooto S, Hangai M, Tomidokoro A, Saito H, Araie M, Otani T, et al Effects of age, sex, and axial length on the three-dimensional profile of normal macular layer structures Invest Ophthalmol Vis Sci. 2011;52:8769–79
24. Kuo AN, McNabb RP, Chiu SJ, El-Dairi MA, Farsiu S, Toth CA, et al Correction of ocular shape in retinal optical coherence tomography
and effect on current clinical measures Am J Ophthalmol. 2013;156:304–11
Source of Support: Nil.
Conflict of Interest: None declared.