Kai Xiong Cheong

Topographic variation of choroidal and retinal thicknesses at the macula in healthy adults.

Background/aims To determine the topographic variation of macular choroidal and retinal thicknesses (RTs) in normal eyes and their relationship with refractive error. Methods Spectral domain optical coherence tomography with enhanced depth imaging was performed on 124 healthy participants using a standardised imaging protocol. Manual segmentation of choroidal boundaries was performed by trained graders, and mean choroidal thickness (CT) was compared with mean RT in corresponding sectors of the Early Treatment Diabetic Retinopathy Study (ETDRS) grid. Results Mean central subfield CT was 322.2 µm. The choroid was thickest at the temporal and superior sectors (323.1–338.1 µm), followed by inferior sectors (314.0–321.8 µm), and thinnest at the nasal sectors (232.8–287.8 µm). In contrast, the retina was thicker nasally (343.4 µm) and thinner temporally (287.1 µm). CT was thickest among emmetropes in all ETDRS subfields and became thinner progressively among low, moderate and high myopes (p<0.001). The variation of both choroidal and RTs among refractive error groups resulted in different topographic patterns at the macula. Conclusion There is significant topographic variation of choroidal and RTs at different regions of the macula, with progressive change of choroidal thickness in all sectors based on the refractive status of the eye.

Comparison of choroidal thicknesses using swept source and spectral domain optical coherence tomography in diseased and normal eyes

Background/aims Choroidal thickness measurements are reported to differ between swept source optical coherence tomography (SS-OCT) and spectral domain OCT (SD-OCT). This study aimed to assess the comparability of choroidal thickness measurements using SS-OCT and SD-OCT devices among patients with retinal diseases and normal participants. Methods In a prospective cohort study of 100 subjects, comprising patients with retinal disease and normal volunteers, OCT scans were performed sequentially with the DRI OCT-1 and Spectralis OCT using standardised imaging protocols. Subfoveal choroidal thicknesses were independently measured by masked reading-centre certified graders. Paired t tests and intraclass correlation coefficients (ICCs) were used to compare the measurements. Results Among all 100 participants, mean subfoveal choroidal thickness was 264.3 µm and 272.4 µm for DRI OCT-1 and Spectralis OCT respectively (p=0.001), with ICC of 0.989. The mean difference in choroidal thickness between OCT devices was larger among eyes with retinal diseases compared with normal eyes (8.4 µm vs 7.3 µm). Eyes with choroidal thickness ≤200 µm had smaller differences between OCT devices compared with those with thicker choroids (mean 3.6 µm vs 10.0 µm, p=0.021). Conclusions Subfoveal choroidal thickness measurements are comparable between DRI OCT-1 and Spectralis OCT. The presence of retinal disease increases the variability of choroidal thickness measurements between OCT devices.

Optical Coherence Tomography Angiography Evaluation of the Parafoveal Vasculature and Its Relationship With Ocular Factors

PURPOSE To determine the size and characteristics of the superficial and deep foveal avascular zone (FAZ) in healthy adults by using optical coherence tomography angiography (OCT-A), and to ascertain the effects of demographic and ocular parameters on the FAZ size. METHODS In a prospective cohort study of 117 healthy volunteers, foveal-centered 3 × 3-mm OCT-A scans were manually graded by certified graders to determine the size of the superficial and deep FAZ. Multiple linear regression analyses were performed to evaluate the impact of demographics and ocular factors, including central retinal thickness (CRT), choroidal thickness, axial length (AL), and spherical equivalent (SE) on superficial and deep FAZ areas. RESULTS The mean age of the participants was 22.5 years, with mean AL of 25.4 mm and mean SE of -4.3 diopters. The mean CRT was 262.8 μm (range, 220-316 μm). The mean superficial FAZ area was 0.24 mm2, while the deep FAZ area was 0.38 mm2 (P < 0.001). Females had a larger superficial (P < 0.001) and deep FAZ (P < 0.001). On univariate linear regression, both superficial and deep FAZ areas had significant correlations with CRT, sex, AL, and SE, but not with age. By multiple linear regression analysis, in normal eyes, superficial FAZ area varied significantly with CRT and sex. Among eyes with high myopia, both superficial and deep FAZ varied significantly with CRT, sex, and choroidal thickness. CONCLUSIONS The superficial and deep FAZ areas varied significantly among healthy eyes. Factors such as CRT, sex, SE, AL, and choroidal thickness influence the size of the FAZ.

Macular Choroidal Thicknesses in Healthy Adults—Relationship With Ocular and Demographic Factors

PURPOSE To determine the differences in choroidal thickness (CT) among different groups of refractive errors and axial lengths, and to describe the rates of change of CT with ocular and demographic factors in various regions of the macula. METHODS Prospective cohort study of 150 healthy volunteers. Spectral-domain optical coherence tomography was performed on both eyes using a standardized imaging protocol. Manual grading of the choroidal boundaries was independently performed by trained graders to determine Early Treatment Diabetic Retinopathy Study (ETDRS) subfield choroidal thickness. Multiple linear regression analyses were performed to determine the effects of spherical equivalent, axial length and age on choroidal thickness in each subfield. RESULTS The mean central subfield CT was 324.9 μm (range, 123-566 μm) and varied significantly with both spherical equivalent (P < 0.001) and axial length (P < 0.001), but not age or sex. On multiple linear regression analysis using spherical equivalent, the coefficients were 20.0 for the central subfield, ranged from 16.9 to 19.9 for the inner subfields, and decreased to 13.9 to 16.2 for the outer subfields. Performing regression analysis using axial length, the coefficients were -36.4 for the central subfield, -30.5 to -34.5 for the inner subfields, and -24.6 to -27.3 for the outer subfields. CONCLUSIONS Choroidal thickness varies significantly with spherical equivalent and axial length in all regions of the macula, but exhibits different rates of change among different subfields. The rates of change were greater in the central and inner subfields compared with the outer subfields.

Comparison of macular choroidal thicknesses from swept source and spectral domain optical coherence tomography.

Background Choroidal thickness (CT) measurements differ between swept source optical coherence tomography (SS-OCT) and spectral domain OCT (SD-OCT) devices for point thickness measurements. We aimed to assess the comparability of mean macular CT measurements between SS-OCT and SD-OCT devices. Methods In a prospective cohort study of 25 healthy volunteers, OCT scans were performed sequentially with the deep range imaging (DRI) OCT-1 and Spectralis OCT using standardised imaging protocols. These OCT scans were independently graded by reading centre-certified graders to obtain mean CT in the various Early Treatment Diabetic Retinopathy Study (ETDRS) subfields. Paired t tests and intraclass correlation coefficients (ICCs) were used to compare the measurements. Results The difference in mean central subfield CT between DRI OCT-1 and Spectralis was 49.3 µm (p<0.001), while differences in CT in various ETDRS subfields varied from 42.1 to 67.2 µm. After manual adjustment of the segmentation boundaries for the central subfield in the DRI OCT-1, the mean central subfield CT for DRI OCT-1 increased from 263.1 to 293.3 µm (p<0.001), and the resultant difference between DRI OCT-1 and Spectralis decreased from 49.3 to 19.1 µm (a decrease of 61.3%; p<0.001). CT between the three-dimensional and radial scanning protocols of the DRI OCT-1 were highly comparable, with differences generally under 10 µm and ICC of 0.888 for the central subfield. Conclusions CT measurements between automated segmentations from the DRI OCT-1 and manual segmentations on the Spectralis OCT may differ by more than 50 µm. This difference can be reduced, but not eliminated, by manual adjustment of segmentation boundaries by trained graders, and should be accounted for when comparing results between the two devices.

Comparison of retinal thicknesses measured using swept-source and spectral-domain optical coherence tomography devices.

BACKGROUND AND OBJECTIVE To compare retinal thicknesses measured using swept-source optical coherence tomography (SS-OCT) and spectral-domain (SD) OCT devices. PATIENTS AND METHODS In a cohort study of 76 healthy eyes and 21 eyes with high myopia, mean retinal thicknesses in ETDRS subfields were compared between OCT scans obtained from the Topcon DRI OCT-1 (Topcon, Tokyo, Japan), Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany), and Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA). RESULTS Central retinal thickness measurements differed significantly among the three OCT devices (Spectralis: 271 µm; Cirrus: 254 µm; DRI OCT-1: 238 µm; P < .001), with mean differences ranging from 15.6 µm to 37 µm. Intraclass correlation coefficients were at least 0.94 for any pair of machines. Similar results were observed in all nine ETDRS subfields. In all sectors, retinal thickness measurements obtained using the 3-D and radial scans of the SS-OCT were similar (mean differences: 0.7 to 3.8 µm). CONCLUSION Retinal thickness measurements obtained from DRI OCT-1 and SD-OCT are different and should be accounted for when comparing results of OCT scans from different devices.

Overestimation of subfoveal choroidal thickness by measurement based on horizontally compressed optical coherence tomography images.

Dear Editor, We read with interest the article by Kim et al. [1] describing differences in manual measurements of choroidal thickness when spectral domain optical coherence tomography (OCT) scans were viewed with various display settings. The authors have made a very important contribution to our understanding of the factors that affect the accuracy of OCTmeasurements. It is now apparent that besides variations in retinal and choroidal measurements that occur between different OCT devices [2–7], important differences can occur within a single device, depending on the image display setting used. In addition to the 1:1 pixel or 1:1 micron display settings that the authors have described in detail, there are other features of the OCT scan that can be performed using the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany) that may alter the relative proportions of the image display. This can in turn affect the accuracy of manual measurements of the choroid and other structures. The Spectralis OCT can perform scans using either the high-speed or high-resolution modes. When the former is used, there are 768 pixels horizontally within each OCT B scan, whereas the high-resolution setting has 1,536 pixels per horizontal scan. Since the number of pixels in the horizontal direction is doubled when the highresolution mode is used, the relative proportions of the OCT scan in the horizontal and vertical directions are correspondingly altered. A high-resolution OCTscan displayed using the 1:1 pixel setting, therefore, will appear to have less horizontal compression compared to a high-speed scan using the same display setting. Figure 1 illustrates displays of OCT scans performed consecutively on the same individual using the high-speed and high-resolution modes. The sections through the fovea were viewed using both 1:1 pixel and 1:1 micron display settings. It is apparent from this comparison that the appearances of the images differ based on both the scan type (Fig. 1a compared with c, and b compared with d) and the display setting (Fig. 1a compared with b, and c compared with d). OCT scans that are performed using the high-speedmode and are viewed with the 1:1 pixel setting will exhibit the greatest amount of horizontal compression of the retinal and choroidal layers compared to OCTscans that are performed using the high-resolution mode, or when the OCT scans are viewed using the 1:1 micron display setting. In the original article [1], Kim et al. cited examples of papers where “the 1:1 micron images were likely used” for choroidal thickness measurements. In fact, for one of the studies cited [8], we can confirm that all OCT scans were performed using the high-resolution mode, and were viewed using the 1:1 pixel C. S. H. Tan :K. X. Cheong Department of Ophthalmology, Tan Tock Seng Hospital, Singapore, Singapore

Change in subfoveal choroidal thickness in central serous chorioretinopathy

Sir, In an article regarding changes in subfoveal choroidal thickness (SFCT) in patients with central serous chorioretinopathy,1 Drs Kang and Kim described decreases in SFCT of 39.9 and 66.9 μm respectively, following observation and treatment with reduced-fluence photodynamic therapy. We would like to highlight some aspects of the study design that may have bearings on the interpretation of these results. Although the authors mentioned diurnal variation in the discussion,1 it does not appear from the description of the methods that this was accounted for in the study design. Earlier studies2, 3 have demonstrated significant diurnal variation of SFCT measured using spectral-domain optical coherence tomography. In these papers, the amplitude (difference between the maximum and minimum choroidal thickness) exceeded 30 μm,2, 3 which is similar in magnitude to the change reported in the observation group in the current paper. Furthermore, when subjects with thicker choroids (defined as ≥400 μm) were sub-analyzed in one paper,2 the mean amplitude was even larger (43.1 μm) with a maximum of 59 μm. In addition to within-subject diurnal variation between the initial and follow-up reviews, it is also important in this study to consider the effects of potentially different measurement times between the two groups of patients (ie, within-group variation), which might have had an effect on the mean choroidal thickness of each group. Another point of interest is whether the authors utilized the eye-tracking feature of the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany) in performing the successive OCT scans between visits. As the choroid is known to exhibit spatial variation in thickness throughout the macula,4, 5 a minor change in the OCT scan position may result in differences in choroidal thickness measurements, which are sufficient to influence the comparison of SFCT. These concerns could have been mitigated by the use of the eye-tracking function that is also available on the Spectralis OCT, which we believe is an important methodological consideration in longitudinal measurements of choroidal thickness. In conclusion, we congratulate the authors on an interesting paper, and urge investigators to consider the impact of diurnal variation of choroidal thickness on the results of such studies.

Changes in choroidal thickness after photodynamic therapy in patients with central serous chorioretinopathy.

Editor, W e read with interest the article by Drs Pryds and Larsen (Pryds & Larsen 2012) describing changes in choroidal thickness after photodynamic therapy (PDT) for central serous chorioretinopathy (CSC). In their article, the authors reported decreases in choroidal thickness averaging 14% and 18% of the pretreatment thickness within the areas of leakage and under the fovea, respectively. These observations are intriguing as they may help enhance our understanding of the pathophysiology of CSC and how PDT affects its clinical course. It would be interesting to know whether the timing of the optical coherence tomography (OCT) scans performed before and after treatment was random, or whether arrangements were made for the patients to be scanned at similar time periods during the day for both the pretreatment and posttreatment OCT scans. Several studies have demonstrated a significant diurnal variation of subfoveal choroidal thickness in healthy adults (Tan et al. 2012; Usui et al. 2012), and it is likely that diurnal variation in the choroidal thickness also occurs in eyes with ocular diseases. Earlier studies have reported amplitudes (the difference between maximum and minimum choroidal thicknesses) ranging from 3 to 67 lm (Tan et al. 2012; Usui et al. 2012). Moreover, the amplitude appears to be greater in eyeswith thicker choroids (Tan et al. 2012). Therefore, it is possible that the variation observed in the authors patients may partly be accounted for by normal diurnal variation in the choroidal vasculature. As pointed out by the authors, one of the strengths of their study is the comparison of choroidal thicknesses between the affected and normal fellow eye. The authors reported that the choroidal thickness was significantly greater in the affected eye compared with the fellow eye prior to the treatment, and this difference decreased to approximately 20 lm after PDT. Table 2 of their manuscript, however, does not show the choroidal thicknesses of the fellow eye both before and after treatment, and this comparison is not reported in the results. It would be interesting to determine the magnitude of the difference in choroidal thickness in the fellow eye before and after treatment. If a significant difference exists, and this difference varies in tandem with the diseased eye for each patient, then this would further support the hypothesis that part of the changemay be due to diurnal variation. An earlier paper (Tan et al. 2012) demonstrated good correlation between changes in choroidal thickness between both eyes of a patient, and one might expect to observe this in these results if the changes are indeed caused by diurnal variation. The authors reported the use of image registration and tracking, so that the subsequent OCT scans were performed at the same locations at the macula. We agree that this is a very important consideration in longitudinal studies of OCT scans, in particular because topographical variations have been reported in choroidal thicknesses throughout the macula (Ouyang et al. 2011). In summary, we congratulate the authors for their interesting and important article. We would like to highlight the importance of accounting for diurnal variation in studies involving longitudinal measurements of choroidal thickness. References

A novel and faster method of manual grading to measure choroidal thickness using optical coherence tomography

PurposeChoroidal thickness (CT) measurements are typically obtained from manual segmentation of optical coherence tomography (OCT) B-scans. This method is time-consuming. We aimed to describe a novel and faster technique to obtain CT measurements.Patients and methodsIn a prospective cohort study of 200 healthy eyes, Spectral-Domain OCT with enhanced depth imaging were performed with the Spectralis OCT using standardised imaging protocols. The OCT scans were independently graded by reading centre-certified graders. The standard method of manual adjustment of segmentation boundaries was performed. The new method consisted of adjusting the lower segmentation line to the choroid-scleral boundary to generate the combined choroid-retina thickness, and subtracting the original retinal thickness (RT) from it to measure CT. Mean CT in the respective Early Treatment Diabetic Retinopathy Study (ETDRS) subfields was measured via the two methods, and were compared with intraclass correlation coefficients (ICC) and Bland–Altman plots.ResultsThe mean central subfield CT was 324.4 μm using the original method, compared with 328.8 μm using the new method, with a mean difference of 4.5 μm (range: −14.0 to +4.0 μm; P<0.001), and ICC for agreement of 0.9996 (P<0.001). Similar comparability was achieved for mean CT across other ETDRS subfields, with mean differences ranging from 2.4 to 3.7 μm, and ICCs ranging from 0.9993 to 0.9995 (all P<0.001).ConclusionsMean CT can be measured by subtracting the original RT from the combined choroid-retina thickness. Only one segmentation line needs to be adjusted, instead of two, reducing time required for segmentation. This method is faster and reliable.