OPTICAL COHERENCE TOMOGRAPHIC ASSESSMENT OF DIABETIC MACULAR EDEMA: COMPARISON WITH FLUORESCEIN ANGIOGRAPHIC AND CLINICAL FINDINGS
Şengül C. Özdek MD, Assistant Professor in Ophthalmology*
M. Alper Erdinç MD, Resident in Ophthalmology*
Gökhan Gürelik MD, Associate Professor in Ophthalmology*
Bahri Aydın MD, Resident in Ophthalmology*
Umut Bahçeci MD, Resident in Ophthalmology*
Berati Hasanreisoğlu MD, Professor in Ophthalmology*
*Gazi University, School of Medicine, Ophthalmology Department, Ankara, Turkey.
Key Words: Diabetic macular edema, optical coherence tomography, fluorescein angiography, visual acuity.
Purpose: To compare the optical coherence tomographic (OCT) features with clinical and fluorescein angiographic (FA) findings in patients with diabetic retinopathy.
Methods: In a retrospective study ophthalmologic examination together with FA’s and OCT images were obtained from 195 eyes of 110 patients with different stages of diabetic retinopathy and OCT images were obtained from 40 eyes of 20 control subjects. Fluorescein leakage characteristics were grouped into five as no-leakage (1), focal (2), diffuse (3), combined focal + diffuse leakage (4) and cystoid (5). Pearson correlation test was used to test the correlation between visual acuity and central foveal thickness and ANOVA was used for the statistical comparison between groups.
Results: The OCT images demonstrated retinal swelling in 66.1% of eyes, cystoid macular edema (CME) in 11.8% of eyes, serous foveal detachment + swelling in 6.2% of eyes, serous foveal detachment + swelling + CME in 3.6% of eyes and normal foveal structure in 12.3% of eyes. The best-corrected visual acuity was significantly correlated with central foveal thickness (R: -0.528, p<0.01). There was 77% agreement between clinical examination and OCT results. CME was detected with OCT in 15.4% of eyes in our study, 40% of which was not detected during slit lamp biomicroscopy and 63.3% of which was not evident in FA. None of the serous foveal detachments could be detected during slit lamp biomicroscopy or FA.
Conclusions: OCT-3 provided objective documentation of foveal structural changes in eyes with diabetic retinopathy. Best-corrected visual acuity provided a significant correlation with the retinal thickness at the central fovea. These results indicate that OCT can facilitate the decision of treatment protocol (surgical or medical) and follow-up of diabetic patients, which will be especially important in the early stages of diabetic maculopathy when the structural changes were not yet evident slit lamp biomicroscopically or angiographically.
Macular edema is a common cause of decreased vision in patients with diabetic retinopathy (DR). Various clinical patterns including focal, diffuse and cystoid macular edema (CME) are observed and different mechanisms are involved in its pathogenesis. Although fluorescein angiography (FA) is used to assess vascular leakage qualitatively in macular edema, actual macular thickening has been shown to be better correlated with visual acuity . With the advent of new technologies, however, quantitative assessment of macular edema has become feasible as determined with retinal thickness. Recent studies have shown a significant correlation between visual acuity and retinal thickness using a retinal thickness analyzer [2, 3] and optical coherence tomography (OCT) [4-12] in eyes with DR.
OCT is a relatively new, non-invasive, non-contact method that allows cross-sectional in vivo imaging of posterior and anterior segment of the eye as well as quantitative measurement of retinal nerve fiber layer. The technique was introduced in 1991 by Huang et al  and has been described in detail by Hee et al . Briefly, the OCT uses low coherence interferometry to provide absolute measurements of retinal thickness and achieves a high axial resolution of approximately 10 µm.
Studies with OCT on diabetic subjects revealed that diabetic macular edema includes three basic structural changes, namely, sponge-like retinal swelling, CME and serous retinal detachment , and mostly macular thickening was suggested to be abnormal when the foveal thickness measures greater than 180µm on OCT [5-11]. Vitreomacular traction which is the result of partial vitreous detachment can only be detected clearly with OCT . In this retrospective study we present the clinical, angiographic and OCT findings in 110 patients with diabetic retinopathy.
PATIENTS AND METHODS
Patients with DR of any stage with clear media without any other ocular diseases were included in the study. Patients underwent a complete ophthalmic examination, including the best-corrected visual acuity measured with Snellen chart, slit lamp biomicroscopy with a +78D lens, intraocular pressure measurement with applanation tonometry, colored fundus photography, FA and OCT examinations. The presence of any proliferative change and clinically significant macular edema (CSME) as defined by the Early Treatment Diabetic Retinopathy Study Group  or CME were assessed clinically by slit lamp biomicroscopy with a +78D lens and noted for each patient. The eyes with no evidence of diabetic maculopathy were also included in the study.
FA was obtained subsequent to injection of 5 ml of 10% sodium fluorescein intravenously. The photographic angle of the fundus camera was 50o. FA’s were assessed according to fluorescein leakage in macular area and grouped into five as no-leakage (1), focal (2), diffuse (3), combined focal + diffuse leakage (4) and cystoid (5). The presence of focal leakage from microaneurysms and dilated capillary segments was named as focal macular edema. The late diffuse leakage of fluorescein from unknown source which is supposed to result from a generalized breakdown of inner blood retinal barrier, leading to leakage from retinal capillaries and even arterioles was categorized as diffuse macular edema. The combined type included focal and diffuse fluorescein leakage together and lastly the late staining of fovea in so-called petaloid appearance that resembles the petals of a flower was named as CME. Patients without any retinal or optic disc neovascularization were grouped as non-proliferative DR group (1) and the remaining was grouped as proliferative DR group (2). The perfusion state of the macula was also assessed as normal (1) or ischemic (2) during FA. Ischemic state of the macula was defined as nonperfusion areas or deformation of the outline of the foveal avascular zone . All angiograms were graded by an experienced vitreoretinal specialist.
OCT was performed using the Humphrey model 3000 (Zeiss-Humphrey Instruments, San Leandro, CA). After pupil dilatation, six consecutive 6mm long scans containing 128 axial profiles (A-scans) at equally spaced angular orientations in a radial spoke pattern centered on the fovea (known as Fast Macular Thickness Protocol) were obtained for each eye. Using Retinal Thickness Mapping Software mean retinal thickness value which was measured in the central disc with a diameter of 1000µm in the center of the macula was used as central foveal thickness. In these cross-sectional images we also noted the presence of foveal traction and structural changes of diabetic macular edema as described by Otani and associates . The structural changes were grouped into 5 as, normal (1), sponge-like retinal swelling (2), CME (3), sponge-like retinal swelling + CME/serous foveal detachment (4) and combination of all types (5).
Agreement rate between the clinical and OCT findings was calculated by dividing the sum of the number of eyes in which macular edema was detected and not detected by both clinical and tomographic examination with the number of all of the studied eyes.
Central foveal thickness measurements were obtained from 40 eyes of 20 age-matched normal subjects (control group) in order to evaluate the normal foveal thickness.
Statistical comparisons between the central foveal thicknesses within fluorescein leakage groups and patterns of macular edema determined by the OCT were performed by using analysis of variance (ANOVA). Pearson correlation coefficient was used to test the correlation between visual acuity and central foveal thickness. Students’ t-test was used to compare the foveal thickness of eyes between foveal perfusion/nonperfusion groups and proliferative DR and non- proliferative DR groups. Statistical analysis was performed using SPSS for Windows (Version 10, SPSS Inc, Chicago, IL, USA).
195 eyes of 110 patients with diabetic retinopathy (54 women and 56 men), ageing between 34 and 77, were included in the study. Visual acuities ranged between 0.05 and 1.0. 148 (75.9%) of 195 eyes were classified as non-proliferative DR and the remaining 47 eyes (24.1%) as proliferative DR after clinical and angiographic examination. CSME was detected clinically in 112 eyes (57.4%) as defined by ETDRS and CME was detected in 18 eyes (9.2%). There was no clinical evidence of macular edema in the remaining 65 eyes (33.3%).
FA demonstrated no dye leakage in 27 eyes (13.8%), focal leakage in 56 eyes (28.7%), diffuse leakage in 56 eyes (28.7%), both focal and diffuse leakage in 45 eyes (23.1%) and CME in 11 eyes (5.6%). The mean central foveal thickness determined with OCT was the least in the no leakage group and was increased in the order of the focal, diffuse and combined leakage groups. Statistical analysis revealed a significant difference between these groups (ANOVA, p=0.000 and Bonferroni correction, p<0.05 for all the groups) (Table 1).
The mean retinal thickness in the central fovea was 355 ± 131 µm (169-754 µm) in study population and 201 ± 20 µm (161-242 µm) in control group ageing between 32 and 73. There was no significant difference between the mean foveal thickness in non-proliferative DR group (348 ± 133) and proliferative DR group (377 ± 125) (t= 0.092, p=0.200).
In the cross-sectional images of OCT, sponge-like retinal swelling was seen solely in 129 (66.1%) eyes which is characterized by increased retinal thickness with reduced intraretinal reflectivity and expanded areas of lower reflectivity (Figure 1). CME was detected solely in 23 (11.8%) eyes which is characterized by intraretinal cystoid spaces in the macular area (Figure 2). Serous foveal detachment was characterized by subretinal fluid accumulation with distinct outer border of the detached retina. Sole serous foveal detachment was not seen in any of the patients. Foveal serous detachment and sponge-like retinal swelling were seen together in 12 eyes (6.2%) and, all patterns were seen together (foveal detachment + sponge-like retinal swelling + CME) in another 7 eyes (3.6%) (Figure 3). There were 24 eyes (12.3%) in which central foveal thickness was less than 200 µm and no evidence of cystoid spaces or subretinal fluid accumulation on tomographic images, hence, we considered these eyes to be normal or non-edematous. There was foveal traction caused by incomplete detachment of posterior hyaloid causing retinal swelling in six eyes (3.1%) (Figure 4) and retinal swelling + serous foveal detachment in one eye (Figure 5).
Retinal swelling was seen in 148 (%75.9) eyes with OCT, 36 (24.3%) of which was not clinically evident as CSME during slit lamp biomicroscopy. CME was detected in 30 eyes (15.4%) as a total with OCT, 12 (40.0%) of which was not clinically evident during slit lamp biomicroscopy and 19 (63.3%) was not detected on FA. OCT showed serous retinal detachment in 19 (9.7%) eyes that was detected neither with slit lamp biomicroscopy nor with FA. Of the 195 eyes, macular edema was detected with both clinical examination and OCT in 128 (66%) eyes, and was not detected by both methods in 22 (11%) eyes. There were 43 (22%) eyes in which macular edema was not detected by clinical examination but was detected by OCT, and only two (1%) eyes in which macular edema was detected by clinical examination but not verified by OCT. Then the agreement rate between the clinical findings and OCT findings was calculated as 0.77 (77%) for the detection of diabetic macular edema.
The mean central foveal thickness was 347 ± 110µm in the sponge like retinal swelling group, 441 ± 96µm in CME group, 460 ± 135µm in sponge like retinal swelling and foveal detachment group and 566 ± 189µm in combined group. Statistical analysis demonstrated a significant difference between these groups (ANOVA, p=0.000, Bonferroni correction, p<0.05 for all patterns).
Central foveal thickness and best-corrected visual acuity showed a significant negative correlation (correlation coefficient: -0.528, p<0.01) (Figure 6). There was no significant difference in foveal thickness between the eyes with macular ischemia and without ischemia (Student’s t-test, t=-0.39, p=0.592) (Table 1). The distribution of structural changes also did not differ between the eyes with ischemia from the others.
Otani et al.  reported that OCT images showed three basic types of macular edema: sponge-like retinal swelling, CME and serous retinal detachment. They have studied patients with CSME and CME, and have reported a significant correlation between the central retinal thickness and visual acuity. In the current study, we have found the similar results in our study population which included even the patients with no ophthalmoscopic evidence of diabetic maculopathy. The mean central foveal thickness was the least in the no-maculopathy group and increased with the order of the only retinal swelling group, CME and combined patterns.
In the current study retinal swelling was the most common tomographic pattern and was seen in 66.1% of the eyes solely and in 76% of eyes together with other patterns. Retinal thickening was not evident clinically as CSME in 24.3% of these eyes. In the cross-sectional images the swollen retina resembled a sponge and had increased retinal thickness, where low reflective areas were expanded and layered structure became irregular. Areas of low reflectivity were prominent in the outer retinal layers. The ratio of eyes with retinal swelling in the present study was similar to Otani and associates’ study (88%) , however, other similar studies reported diffuse retinal swelling without cystoid appearance in 40-55% and CME in 30-60% of eyes with diabetic retinopathy [6,11-12].
CME was detected with OCT in 30 (15.4%) eyes in our study, 40% of which was not detected during slit lamp biomicroscopy with a +78D lens and 63.3% of which was not noted during angiography. This is a striking finding, demonstrating the importance of OCT in the detection of foveal diabetic changes which are not evident clinically and angiographically. The great discrepancy between FA and OCT could be also because of masked cystoid staining pattern in eyes with severe focal and diffuse leakage. Slit-lamp biomicroscopy with +78D lens is a reliable method for the assessment of macular edema, however, since this was a retrospective study, some of the cases may have been overlooked which might increase the percentage of eyes with macular edema missed clinically. OCT seems to be very useful and easy to use device for the assessment of the type of diabetic maculopathy and to plan the treatment protocol.
We have demonstrated serous foveal detachment in 9.7% of cases which was not detected during clinical examination. This is also very important finding which may explain the cause of unexplained visual loss in patients with diabetic maculopathy. Serous foveal detachment determined with OCT was reported in 0-15% of diabetic patients in the literature [6, 11, 12].
Recently Browning et al.  have compared the clinical examination with a +78 D non-contact lens and OCT in the assessment of DME and reported that the raw agreement of the 2 methods were 0.83 for the foveal region. In our study, the raw agreement can be calculated as 0.77 (77%) for the 2 methods when calculated in the same way as they described (18). In our study we used a non-contact +78 D lens, however contact type lenses are considered superior to +78 D non-contact lens for examining macula in more detail. In a prospective study by Brown et al.  comparing contact lens biomicroscopy with OCT for the detection of diabetic foveal edema, overall agreement between the results of the 2 method was 69%, which was only 23% in cases with mild foveal thickening and 85% when the eyes with mild foveal thickening were excluded. All of these studies demonstrate the value of OCT for the detection of foveal thickening especially when it is mild which can easily be overlooked clinically.
Kaiser et al.  demonstrated a localized, shallow, subclinical, tractional macular detachment in eight of nine patients with DME associated with posterior hyaloid traction. Imai et al.  who studied eyes with proliferative DR showing tractional macular elevation, observed retinoschisis with or without associated retinal detachment in 16 of 17 eyes and retinal detachment in only six of 17 eyes. Kang et al.  reported foveolar detachment with apparent vitreofoveal traction in 2.8% of eyes with DME. In our study population we have demonstrated posterior hyaloidal traction causing retinal swelling or serous foveal detachment in 3.6% of the study eyes. Previous reports in the literature documented improvement in visual function after pars plana vitrectomy with separation and removal of the posterior hyaloid in eyes with DME associated with posterior hyaloidal traction [22, 23]. OCT remains the mainstay for the decision of surgery by demonstrating the vitreofoveal traction which is very difficult to demonstrate by any other method in diabetic maculopathy.
FA is known to be a sensitive method for the qualitative assessment of fluid leakage in diabetic macular edema, however, actual macular thickening is better correlated with loss of visual acuity. Furthermore, FA is an invasive test, with side effects ranging from nausea in up to 20% of cases to its rare complications of anaphylaxis and death [24, 25]. The information it provides is qualitative and the interpretation of the results may be subjective. OCT is noninvasive, comfortable, safe and fast and can be repeated as often as is required and offers an alternative to the FA in follow-up of changes in retinal thickness after laser photocoagulation and intravitreal steroid injections. However, FA is still essential for the assessment of foveal perfusion state which can not be demonstrated with OCT. After an initial FA, OCT seems to be a useful non invasive tool in close follow-up of the effectiveness of treatment modalities in diabetic maculopathy.
In conclusion, this study has demonstrated that OCT provides an objective documentation of foveal structural changes in eyes with diabetic retinopathy. The best-corrected visual acuity was significantly correlated with central foveal thickness (measured with OCT) in these patients. Although there is a good agreement between clinical examination and OCT, some of the diabetic structural changes in fovea such as foveal traction, serous foveal detachment and CME can be detected with OCT, which may not be evident in ophthalmoscopy and FA.
1- Nussenblatt RB, Kaufman SC, Palestine AG, Davis MD, Ferris FL: Macular thickening and visual acuity. Measurement in patients with cystoid macular edema. Ophthalmology 1987;94:1134-1139.
2- Shahidi M, Ogura Y, Blair NP, Rusin MM, Zeimer R: Retinal thickness analysis for quantitative assessment of diabetic macular edema. Arch Ophthalmol 1991;109:1115–1119.
3- Yasukawa T, Kiryu J, Tsujikawa A, Dong J, Takagi H, Ogura Y: Quantitative analysis of foveal retinal thickness in diabetic retinopathy with the scanning retinal thickness analyzer. Retina 1998;18:150 –155.
4- Hee MR, Puliafito CA, Wong C, Duker JS, Reichel E, Rutledge B, Schuman JS, Swanson EA, Fujimoto JG: Quantitative assessment of macular edema with optical coherence tomography. Arch Ophthalmol 1995;113:1019 –1029.
5- Hee MR, Puliafito CA, Duker JS, Reichel E, Coker JG, Wilkins JR, Schuman JS, Swanson EA, Fujimoto JG: Topography of diabetic macular edema with optical coherence tomography. Ophthalmology 1998;105:360 –370.
6- Otani T, Kishi S, Maruyama Y: Patterns of diabetic macular edema with optical coherence tomography. Am J Ophthalmol 1999;127:688–693.
7- Sanchez-Tocino H, Alvarez-Vidal A, Mandonado MJ, Moreno-Montanes J, Garcia-Layana A: Retinal thickness study with optical coherence tomography in patients with diabetes. Invest Ophthalmol Vis Sci. 2002;43:1588-1594.
8- Goebel W, Kretzchmar-Gross T: Retinal thickness in diabetic retinopathy: a study using optical coherence tomography (OCT). Retina. 2002;22:759-67.
9- Yang CS, Cheng CY, Lee FL, Hsu WM, Liu JH: Quantitative assessment of retinal thickness in diabetic patients with and without clinically significant macular edema using optical coherence tomography. Acta Ophthalmol Scand. 2001;79:266-270.
10- Lattanzio R, Brancato R, Pierro L, Bandello F, Iaccher B, Fiore T, Maestranzi G: Macular thickness measured by optical coherence tomography (OCT) in diabetic patients. Eur J Ophthalmol. 2002;12:482-487.
11- Yamamoto S, Yamamoto T, Hayashi M, Takeuchi S: Morphological and functional analysis of diabetic macular edema by optical coherence tomography and multifocal electroretinograms. Graefe’s Arch Clin Exp Ophthalmol. 2001;239:96-101.
12- Kang SW, Park CY, Ham DI: The correlation between fluorescein angiographic and optical coherence tomographic features in clinically significant diabetic macular edema. Am J Ophthalmol 2004;137:313-22.
13- Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, Hee MR, Flotte T, Gregory K, Puliafito CA, Fujimoto JG : Optical coherence tomography. Science 1991;254:1178–1181.
14- Hee MR, Izatt JA, Swanson EA, Huang D, Schuman JS, Lin CP, Puliafito CA, Fujimoto JG: Optical coherence tomography of the human retina. Arch Ophthalmol 1995;113: 325-332.
15- Hee MR, Puliafito CA, Wong C, Duker JS, Reichel E, Schuman JS, Swanson EA, Fujimoto JG: Optical coherence tomography of macular holes. Ophthalmology. 1995;102:748-756.
16- Early Treatment Diabetic Retinopathy Study Research Group. ETDRS Report Number 1. Photocoagulation for diabetic macular edema. Arch Ophthalmol 1985;103:1796–1806.
17- Early Treatment Diabetic Retinopathy Study Research Group. ETDRS Report Number 11. Classification of Diabetic Retinopathy from Fluorescein Angiograms. Ophthalmology 1991;98:807-22
18- Browning DJ, McOwen MD, Bowen RM Jr, O'Marah TL: Comparison of the clinical diagnosis of diabetic macular edema with diagnosis by optical coherence tomography. Ophthalmology 2004;111:712-5.
19- Brown JC, Solomon SD, Bressler SB, Schachat AP, DiBernardo C, Bressler NM: Detection of diabetic foveal edema: contact lens biomicroscopy compared with optical coherence tomography. Arch Ophthalmol 2004;122:330-5.
20- Kaiser PK, Riemann CD, Sears JE, Lewis H: Macular traction detacment and diabetic macular edema associated with posterior hyaloidal traction. Am J Ophthalmol 2001;131:44–49.
21- Imai M, Iijima H, Hanada N: Optical coherence tomography of tractional macular elevations in eyes with proliferative diabetic retinopathy. Am J Ophthalmol. 2001;132: 458-461.
22- Lewis H, Abrams GW, Blumenkranz MS, Campo RV: Vitrectomy for diabetic macular traction and edema associated with posterior hyaloidal traction. Ophthalmology 1992;99:753–759.
23- Massin P, Duguid G, Erginay A, Haouchine B, Gaudric A: Optical coherence tomography for evaluating diabetic macular edema before and after vitrectomy. Am J Ophthalmol 2003;135:169–177.
24- Stein MR, Parker CW. Reactions following intravenous fluorescein. Am J Ophthalmol 1971;72:861– 8.
25- Schatz H, Farkos WS. Nausea from fluorescein angiography. Am J Ophthalmol 1982;93:370–1.
Table 1: Central foveal thickness (OCT) and FA characteristics
FA characteristics Mean Foveal thickness
No leakage 267 ± 87µm
Focal leakage 306 ± 116µm
Diffuse leakage 385 ± 116µm
Focal + diffuse leakage 415 ± 146µm
CME 422 ± 109µm
Ischemia (+) 366 ± 128µm
Ischemia (-) 353 ± 132µm
Figure 1: An OCT picture for sponge-like retinal swelling. Central foveal thickness is 486 µm.
Figure 2: An OCT picture for CME
Figure 3: An OCT picture for serous foveal detachment + sponge-like retinal swelling + CME.
Figure 4: An OCT picture for retinal swelling caused by posterior hyaloidal traction.
Figure 5: An OCT picture for serous foveal detachment + sponge-like retinal swelling caused by posterior hyaloidal traction.
Figure 6: Box plot of central macular thickness and visual acuity. A significant negative correlation (correlation coefficient: -0.528, p<0.01) is seen between the retinal thickness at the fovea and best corrected visual acuity. The bars represent the 25%-75% percentiles.