Short wavelength automated perimetry and standard automated perimetry in central serous chorioretinopathy

Short wavelength automated perimetry (SWAP) is known for detecting the early reduction of retinal sensitivity (RS) in glaucoma. It’s application in retinal diseases have also been discussed previously. We investigated the difference in RS measured between standard white-on-white automated perimetry (WW) and blue-on-yellow SWAP in central serous chorioretinopathy (CSC). The overall RS (W-RS, S-RS) as well as the RS inside and outside of the serous retinal detachment (SRD) region were investigated in 26 eyes of 26 CSC patients using WW and SWAP. The central retinal thickness, central choroidal thickness, SRD area (SRDa), and SRD height at the fovea were measured using optic coherence tomography. RS inside the SRD region was lower than that of outside for both perimetries (both p < 0.001). The difference between RS inside and outside of the SRD region was greater in SWAP compared to WW (p < 0.001). Univariate analysis revealed significant correlations between SRDa and both W-RS and S-RS (both p < 0.001); moreover, multivariate analysis indicated that only S-RS was selected as the optimal model for SRDa. Our study demonstrated that SWAP was detected the decrease in RS more accurately than WW in CSC. These results may suggest the usefulness of SWAP for detecting change of retinal function in CSC.

The optimal models for logMAR BCVA, W-RS, and S-RS, as a result of multivariate analysis using Akaike's information criterion (AIC) model selection, are shown below:

Discussion
In the present study, RS was measured using SWAP and WW perimetric tests in 26 CSC eyes with SRD, along with the SRD measurements with OCT. As a result, it was suggested that the difference of the retinal sensitivities within and outside the SRD region was significantly larger with SWAP than with WW perimetry. Furthermore, the optimal model for SRDa only included S-RS but not W-RS. These outcomes suggest the usefulness of SWAP in the evaluation of visual functions in CSC eyes compared to WW perimetry. Previous studies have shown the usefulness of SWAP in evaluating retinal functions in some retinal diseases such as diabetic retinopathy [26][27][28][29][30][31][32][33][34][35][36][37][38][39] , age-related macular degeneration [40][41][42][43] , birdshot retinochoridopathy 44,45 , inherited retinal degeneration 46 , idiopathic optic neuropathy 47 , Afrashi et al. has investigated the change in the central RS in 18 resolved CSC patients and compared to that in normal controls, using SWAP and WW perimetries 25 . As a result, the mean deviation value was significantly lower in the CSC group compared with the control group for both SWAP and WW perimetry, despite the preserved VA (20/20) in all subjects with CSC. With SWAP, other perimetric parameters (pattern standard deviation, short-term fluctuation, and corrected pattern standard deviation) were also significantly lower in the CSC group than in the control group, which was not the case with WW perimetry. This result suggests the potential usefulness of SWAP in eyes with CSC over WW perimetry; however, much more detailed investigation is required, as this study merely compared the overall RSs in the measured VF. In contrast, the current study investigated RSs inside and outside the SRD-affected region separately both with SWAP and WW perimetries. In general, RS decreases as the eccentricity from the fovea increases 48,49 . Nonetheless, in the current study, W-RS_in was significantly lower than W-RS_out agreeing with previous studies (Fig. 2a) 15,50 . Similar tendency was observed in the comparison between S-RS_in and S-RS_out (Fig. 2b). It should be postulated that the difference between S-RS_in and S-RS_out was significantly larger than between W-RS_in and W-RS_out (Fig. 2c). This cannot be fully explained by the difference of W-RS and S-RS, since S-RS was significantly lower than W-RS. Indeed an additional experiment using the difference rate relative to W-RS and S-RS yielded a significant large difference rate with SWAP (data not shown in Result).
In our previous study, we investigated the association between macular volume of SRD and the RS at 0, 2, 4, and 6 degrees from the fovea, which suggested that RS was significantly associated with macular volume at the   www.nature.com/scientificreports/ corresponding region, and not at other regions 15 . LogMAR BCVA mainly reflects the retinal function around the fovea only, whereas RS is the result of the retinal function in a wider area. Agreeing with this, in the current study, SRDh was included in the optimal model for logMAR BCVA but not in those for W-RS and S-RS. In contrast to SRDh, SRDa is a structural parameter obtained from a wider region of the retina, and indeed, SRDa was included in the optimal models for both W-RS and S-RS. This is in agreement with a previous study by Sekine et al. in that a significant decrease of RS was observed in the SRD-affected region 52 . Of note, it should be postulated that W-RS was not included in this optimal model for SRDa, which suggests that SWAP was more useful than WW perimetry when analyzing the structure-function relationship using SRDa. SWAP selectively evaluates the function the blue cone pathway by projecting a blue stimulus on a yellow background, which hyperpolarizes the rods as well as other long-and middle-wavelength-sensitive cones 16,51,52 . Only about 5-10% of the cone cells in the retina are blue cones [10][11][12][13][14][17][18][19][20][21][22]51 . In particular, they are distributed less in number around the fovea compared to the surrounding retina 48 . In glaucoma, this sparse distribution of blue cone pathways has been reported to enable early detection of VF damage 13,18 . For instance, Johnson et al. and Sample et al. have reported that SWAP deficits are predictive of impending glaucomatous VF loss for standard WW perimetry in their longitudinal studies using ocular hypertensive and glaucomatous eyes 13,18 . Additionally, Machida et al. suggested that blue cone pathways may be more delicate and susceptible to photoreceptor diseases 53 . The response range of these cones is also significantly limited compared with the other cones, hence an equivalent loss of function among the cone cells would result in a larger response in the blue cone pathway 53 .
A limitation of the current study was that it was not possible to measure the foveal sensitivity using SWAP. Such measurement will enable further detailed investigation of the association with SRDh and visual function with SWAP and WW perimetry. Moreover, it may be possible for some regions of the retina outside of the SRD region to have locally decreased RS as they may indicate areas of resolved SRD from the past, retinal atrophy, or retinal pigment epithelium cracks. Since this was a cross-sectional study, we did not follow up on the disease duration and clinical course of each subject. It is of our interest to further investigate using image tools, such as fundus autofluorescence, to highlight such atrophic regions in areas outside of the current SRD and assess their retinal sensitivity difference.
In conclusion, our study demonstrated that the difference between RS outside and inside of the SRD region was greater in SWAP than WW, suggesting that SWAP is more sensitive in detecting retinal dysfunction in CSC eyes. Furthermore, SRDa was better correlated with W-RS than W-RS, suggesting that SWAP is more sensitive to morphological change in CSC eyes. These results suggest the usefulness of SWAP for evaluating the retinal function as a result of structural changes of the retina in CSC patients.

Methods
This study was approved by the Research Ethics Committee of the Graduate School of Medicine and Faculty of Medicine at The University of Tokyo (Approval ID: 3770). Informed consent was obtained from all participating subjects. This study was performed according to the tenets of the Declaration of Helsinki.
Cross-sectional case series was employed in the current study. Twenty-six eyes of 26 CSC patients (20 males and 6 females) with intra and/or extra foveal SRDs who visited The University of Tokyo Hospital were included as subjects of this study. All patients underwent comprehensive ophthalmological examinations including measurement of BCVA and intra-ocular pressure, anterior segment examination, and fundoscopy under pupil dilation. The diagnosis of CSC was based on findings from OCT, fluorescein angiography, and indocyanine green angiography. All subjects underwent OCT imaging, as well as SWAP and WW perimetric tests on the same day. Eyes with other ocular diseases such as age-related macular degeneration and choroidal neovascularization were carefully excluded.
BCVA was measured using the Landolt chart and were converted to logMAR BCVA for analysis. Spectral domain OCT images were obtained using the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). All OCT images consisted of line scans (horizontal and vertical B-scans) and raster scans (25 horizontal B-scans). CRT, CCT, and SRDh were measured using images obtained from enhanced depth imaging (EDI) mode. SRDa was measured by evaluating each of the 25 B-scans for SRD; the edges of the SRD were marked on each scan. All markers indicating the edge of the SRD were then connected to identify the SRD-affected area. If more than one isolated SRD region existed in the eye, SRDa was calculated as the area sum of all observed SRD regions.
Both SWAP and WW perimetric tests were performed using the AP-7000 automatic perimeter (KOWA Company Ltd., Tokyo, Japan) using the full-threshold strategy (Goldmann size III). RS thresholds were measured within 10-degree radius from the fovea, corresponding to the Humphrey Field Analyzer 10-2 program (Carl Zeiss Meditec, Dublin, CA). SWAP was conducted using a 450 nm blue narrow band target which was presented for 200 ms, similarly to the WW perimetric test. The target was projected on a yellow background (600 nm with a luminance of 100 cd/m 2 ). The order of SWAP and WW was performed randomly in order to minimize the learning effect of perimetry. Fixation point was checked using MP-3 microperimetry (Nidek Co. Ltd., Aichi, Japan) in all subjects.
The SWAP and WW perimetry results were superimposed on the corresponding OCT images including the marked SRD region using the built-in software in the AP-7000 automatic perimeter (Fig. 3).
W-RS and S-RS were calculated. In addition, the mean RS inside (W-RS_in, S-RS_in) and outside (W-RS_out, S-RS_out) of the SRD-affected regions were also calculated.

Statistical analysis.
Comparisons were conducted (i) between W-RS_in and W-RS_out, (ii) between S-RS_in and S-RS_out, (iii) between W-RS_in and S-RS_in, and (iv) between W-RS_out and S-RS_out, using the exact Wilcoxon signed rank test. Subsequently, the differences between W-RS_in and W-RS_out and also www.nature.com/scientificreports/ between S-RS_in and S-RS_out were calculated and compared to each other using the exact Wilcoxon signed rank test. Additionally, for each of the visual function parameters (logMAR BCVA, W-RS, and S-RS), the association with age and the four structural parameters (CCT, CRT, SRDh, and SRDa) was analyzed using univariate and multivariate linear regressions. Following the multivariate linear regression, model selection was performed to identify the optimal linear regression models using the second-order bias-corrected AICc index from all 2 5 patterns consisting of five variables (age, CCT, CRT, SRDh, and SRDa). The AIC is a well-known statistical measurement used in model selection, and the AICc is a corrected version of AIC, which provides an accurate estimation even when the sample size is small 54,55 . The selected variables through the model selection were regarded as statistically significant. Similarly, the association between SRDa and visual function parameters (logMAR BCVA, W-RS, and S-RS) was analyzed and the optimal model was identified.
All statistical analysis was performed using the statistical programming language "R" (R version 3.5.1; The R Foundation for Statistical Computing, Vienna, Austria).

Data availability
The datasets generated and analyzed during the current study are available from the first/corresponding authors on reasonable request.