Patients

One hundred twenty-five eyes of 73 patients (37 men and 36 women; mean age, 65.2 ± 9.6 years; range, 34-89 years) were referred to the Department of Ophthalmology in Kyoto University Hospital from January 2008 to February 2010 and studied retrospectively. Two patients had diabetes mellitus (DM) type 1 and 71 DM type 2 with DR; 5 eyes had mild nonproliferative diabetic retinopathy (NPDR), 44 moderate NPDR, 51 severe NPDR, and 25 proliferative diabetic retinopathy. Eyes that had not been treated for DME and had OCT images of sufficient quality were consecutively included.

Qualitative and Quantitative Assessment Using Spectralis OCT

After fundus biomicroscopy and measurement of the best-corrected VA (BCVA), retinal sectional imaging using Spectralis OCT was delineated, with infrared fundus imaging. For qualitative and quantitative analysis of OCT images, 20-degree radial scans centered on the fovea were obtained in a clockwise manner, and 49 raster scans were used to evaluate the mean retinal thickness (from the innermost internal limiting membrane [ILM] to the retinal pigment epithelium [RPE]) at the fovea (radius, 500 μm) and 4 subfields (superior, nasal, inferior, and temporal) in the parafoveal areas (radius, 500-1500 μm) according to the manufacturer’s protocol.

Spectralis OCT enabled clear visualization of the individual layers; the presumed fovea first was defined as the central area without the inner retinal layers, NFL, GCL, IPL, and INL, whether retinal detachment, cystoid spaces, or retinal swelling was present or absent. All or any of these layers, which were well delineated in the periphery, were traced to the central area until the area where these layers disappeared was determined. We next used a modified classification of the morphology at the presumed fovea to classify the pathology into 3 groups: the SRD type had SRD but not cystoid spaces at the presumed foveal center; the CME type had predominantly foveal cystoid spaces and was sometimes accompanied by a minimal SRD; and the diffuse type had neither SRD nor cystoid spaces at the presumed fovea. The center point thickness (CPT), the vertical length between the innermost retina and the RPE, was measured as described previously. In eyes with the SRD type, we manually measured the height of the SRD (the vertical length from the outermost sensory retina to the RPE at the presumed fovea) and the thickness of the sensory retina at the fovea (the vertical length from the innermost ILM to the outermost outer segments). In eyes with the CME type, the height of the cystoid space (the vertical length from the innermost to the outermost of cystoid spaces) and the outer retinal thickness (the vertical length from the base of the cystoid space to the RPE) at the fovea were measured manually.

We qualitatively evaluated the status of the ELM at the presumed fovea as representative of the cone photoreceptors. The ELM is seen as a hyperreflective line over the IS/OS in Spectralis OCT images. We defined a well-delineated and continuous ELM as intact; otherwise it was considered to be disrupted. To assess the bipolar cells, which transmit light stimulation in the cone photoreceptors at the fovea, we evaluated the retinal cystic changes horizontally 500 μm from the presumed foveal center. We assumed that the area around that point contained the bipolar cells, which receive the projection from the cone photoreceptors at the fovea, because histologic reports have described that the secondary or tertiary neurons corresponding to cone photoreceptors at the physiologic fovea are displaced centrifugally to 300 to 450 μm. In addition, pathohistologic publications have reported that cystoid spaces are mainly in the INL and OPL in ME associated with retinal vascular diseases, and Spectralis OCT also showed cystic changes mainly in the INL and OPL/ONL. Therefore, we estimated the cystic changes in these layers. Briefly, the presence of cystic changes was evaluated at each clock hour in radial sections and scored from 0 (absent at any clock hour) to 12 (present at all clock hours), and the correlation with the logarithm of minimal angle of resolution (logMAR) VA was calculated by Spearman rank correlation coefficient. We did not divide the OPL and ONL on Spectralis OCT. Although the OPL in the parafovea is much thicker than the ONL in histologic sections, Spectralis OCT often shows a thinner OPL, suggesting that this instrument shows only a part of the histologic OPL as a hyperreflective layer. Most cystoid spaces in the OPL/ONL are seen in the OPL or the inner side of the ONL on Spectralis OCT, which might correspond to those in the OPL in pathohistologic reports.

Statistical Methods

The results are expressed as the mean ± standard deviation (SD). The Student t test was used to compare quantitative data populations with normal distributions and equal variance. The data were analyzed using the Mann-Whitney U test and the Kruskal-Wallis test for populations with non-normal distributions or unequal variance. Linear regression analysis or Spearman rank correlation coefficient was performed to test the statistical correlation. A P < .05 was considered statistically significant.

Foveal Cystoid Spaces

Among several cellular components affecting the VA in the retina, we first evaluated the status of the cone photoreceptors at the fovea and the correlation with the logMAR VA in eyes with the CME type ( Figure 1 ). Nine eyes with the CME type had intact ELM at the presumed fovea ( Figure 1 ) and better logMAR VA than 11 eyes without intact ELM (0.262 ± 0.180 vs 0.621 ± 0.287), although the outer retinal thickness, where the cone photoreceptors might be, was not correlated with the logMAR VA. The CPT, the mean thickness at the center (radius, 500 μm), and the height of the cystoid spaces were significantly correlated with the logMAR VA ( Table 1 ). The mean parafoveal thickness and cystic changes was more significantly correlated with the logMAR VA ( Table 1 ).

Foveal cystoid spaces in diabetic retinopathy. The representative images of foveal cystoid spaces in a 20-degree Spectralis OCT scan (Top right) and corresponding infrared images (Top left). Nine eyes had an intact external limiting membrane (ELM) at the presumed fovea (Middle left; arrowhead), whereas 11 eyes did not (Middle right; arrowhead). (Bottom left) Cystoid spaces 500 μm from the foveal center. (Bottom right) There are no apparent cystoid spaces at the same point despite the definite foveal cystoid space. The arrows indicate the presumed foveal center. The arrowheads indicate 500 μm from the foveal center.

Serous Retinal Detachment at the Fovea

We investigated the pathologic changes at the fovea and the parafovea in eyes with the SRD type ( Figure 2 ). Spectralis OCT showed intact ELM at the presumed fovea in 17 eyes, and the logMAR VA in those eyes did not differ significantly from that in eyes in which the ELM was disrupted (0.211 ± 0.192 vs 0.269 ± 0.105) ( Figure 2 ). We analyzed the height of the SRD and the thickness of the sensory retina at the presumed fovea and found that these parameters were not correlated with the logMAR VA ( Table 2 ). Both qualitative and quantitative analyses in the parafoveal areas showed no association between each parameter and logMAR VA.

Serous retinal detachment at the fovea in diabetic retinopathy. The representative images of foveal serous retinal detachment in a 20-degree Spectralis OCT scan (Top right) and corresponding infrared images (Top left). Seventeen eyes had an intact external limiting membrane at the presumed fovea (Bottom left; arrowhead), whereas 4 eyes did not (Bottom middle left; arrowhead). (Bottom middle right) Cystoid spaces are seen 500 μm from the foveal center. (Bottom right) There are no apparent cystoid spaces at the same point. The arrows indicate the presumed foveal center. The arrowheads indicate 500 μm from the foveal center.

Absence of Serous Retinal Detachment or Cystoid Spaces at the Fovea

In 84 eyes with the diffuse type, intact ELM was identified at the presumed fovea in 70 eyes, which had significantly better logMAR VA than those in which the ELM was disrupted (0.066 ± 0.157 vs 0.626 ± 0.273) ( Figure 3 , Table 3 ). The CPT and the mean central thickness were positively correlated with the logMAR VA ( Table 3 ). Furthermore, the increased thickness in the parafovea and cystic changes 500 μm from the foveal center also were significantly associated with poor logMAR VA ( Table 3 ).

The absence of cystoid spaces or serous retinal detachment at the fovea in diabetic retinopathy. The representative images of retinas with no foveal cystoid spaces or serous retinal detachment in a 20-degree Spectralis OCT scan (Top right) and corresponding infrared images (Top left). Seventy-nine eyes had an intact external limiting membrane (ELM) at the presumed fovea (Middle left; arrowhead), whereas 25 eyes had no or a partially distinct ELM (Middle right; arrowhead). (Bottom left) Cystoid spaces 500 μm from the foveal center are seen. (Bottom right) No apparent cystoid spaces are seen at the same point. The arrows indicate the presumed foveal center. The arrowheads indicate 500 μm from the foveal center.

TABLE 3

Correlation Between Logarithm of Minimal Angle of Resolution and Each Parameter in 84 Eyes With No Cystoid Spaces or Serous Retinal Detachment at the Presumed Fovea

Parameter		Correlation With logMAR
ELM	Intact, 70 eyes	P < .001
	Disrupted, 14 eyes
CPT (μm)	282 ± 73	P < .004, R = .315
Mean thickness in each subfield (μm)
Center	325 ± 63	P < .001, R = .381
Parafoveal area
Superior	379 ± 56	P = .002, R = .338
Nasal	368 ± 40	P = .293, R = .116
Inferior	366 ± 47	P < .001, R = .395
Temporal	370 ± 68	P < .001, R = .532
Average	371 ± 46	P < .001, R = .426
Cystic changes in parafoveal area
INL alone		P < .001
OPL/ONL alone		P = .009
Both INL and OPL/ONL		P < .001

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. In Vivo Morphologic Characteristics of Salzmann Nodular Degeneration With Ultra-High-Resolution Optical Coherence Tomography
2. Comparison of the Correlations Between Optic Disc Rim Area and Retinal Nerve Fiber Layer Thickness in Glaucoma and Nonarteritic Anterior Ischemic Optic Neuropathy
3. A Reappraisal of Eyelid Eccrine and Apocrine Hidrocystomas: Microanatomic and Immunohistochemical Studies of 40 Lesions
4. Book review

Stay updated, free articles. Join our Telegram channel

Join