**4. Optical coherence tomography based classification of macular edema**

OCT is a sensitive tool to diagnose, quantify, and classify diabetic macular edema. The first OCT-based classification for DME was given by Otani et al [15]. They were the first to identify 3 patterns of fluid accumulation, including sponge-like retinal swelling, cystoid macular edema, and serous retinal detachment (**Figure 3**). They further described that early changes of macular edema were confined to the outer retinal layer mainly the outer plexiform layer when compared to histopathology [15]. With the further accumulation of fluid, the inner retinal layers were involved. The presence of serous retinal detachment in patients with DME is a finding which may not be easily distinguished on biomicroscopy or FA.

In 2004, Panozzo proposed a classification system based on five parameters: retinal thickness, volume, morphology, diffusion, and presence or absence of vitreoretinal traction [16]. They quantified the retinal thickness and volume in three different zones around the fovea. The types of macular edema observed were in agreement with that described by Otani et al., [15] with the only difference being that the size of the cyst was measured to subclassify the grade of the cystoid variety of macular edema. The presence of epiretinal traction and its pattern (tangential or

**Figure 3.** *Cystoid macular edema with presence of serous retinal detachment (spectral-domain OCT).*

anteroposterior) were also described. This distinguished cases with an additional component of retinal distortion (**Figure 4**). In 2006, Kim et al. demonstrated similar findings of macular edema and posterior hyaloid traction. In addition, they described tractional retinal detachment as a peak-shaped detachment of the retina [17]. These 3 previous classifications used TD-OCT (**Table 1**).

With the advent of SD-OCT, Murakami et al. for the first time showed that in addition to the morphology of edema, the photoreceptor status played a significant role in the prognosis of visual acuity [18]. They classified edema into serous retinal detachment, cystoid macular edema, and Diffuse type (absence of either cystoid macular edema or serous retinal detachment) with the latter term being used for cases that had retinal thickening but an absence of cysts or serous fluid [18]. Later in 2012, Koleva-Georgieva proposed a classification in which the term early subclinical macular edema was introduced, to describe cases with macular edema which were previously being missed on clinical examination [19]. In addition, they also included the integrity of both the outer retinal layers, the IS/OS (inner segmentouter segment junction, now identified as the EZ layer), and the ELM. Retinal morphology, topography, and presence of traction at macula were also a part of the classification and were similar to the other classifications [19]. In 2013, Helmy et al. further subclassified cystoid macular edema based on the proportion of the largest cyst to the maximum retinal thickness(CME Grade I-IV). The integrity of IS/OS

**Figure 4.** *Vitreomacular traction in a case of diabetic macular edema (captured with SD-OCT).*


#### **Table 1.**

*Time domain-OCT and Spectral domain-OCT based classification of Diabetic macular edema.*

junction and ELM, presence or absence of neurosensory detachment, or vitreoretinal traction were also included. They extended their classification to include the presence of hyperreflective foci in the outer retina from the ELM to the RPE [20].

#### **5. Role of OCT in treatment of diabetic macular edema**

The introduction of intravitreal anti-vascular endothelial growth factor (anti-VEGF) agents significantly changed the treatment of DME a few years ago [22, 23]. Though laser treatment prescribed by the ETDRS study reduced the risk of vision loss significantly, only 20% of laser-treated eyes experienced a gain in visual acuity of at least 3 lines (15 letters) at 2 years [24]. A study by DRCR.net compared the efficacy of anti-VEGF treatment with laser treatment in eyes with DME [25, 26]. Results showed that anti-VEGF therapy was more effective in preventing the loss of visual acuity. In addition, a significant percentage of eyes showed an improvement in mean visual acuity [25, 26].

*Optical Coherence Tomography in Diabetic Retinopathy DOI: http://dx.doi.org/10.5772/intechopen.100587*

Monthly injections and follow-up with OCT imaging of the macula have been recommended in various guidelines [27–30]. Monthly treatment till there is no edema on follow-up OCT scan and reinitiating treatment when edema recurs or vision deteriorates is the preferred clinical practice for the management of DME [30, 31].

However, according to the FDA label of Eylea® (aflibercept), 'the recommended dose for eylea (for DME) is 2 mg (0.05 mL) administered by intravitreal injection every 4 weeks (approximately every 28 days, monthly) for the first 5 injections followed by 2 mg (0.05 mL) via intravitreal injection once every 8 weeks (2 months)' [32].

Cases that do not show a response after 3 monthly injections are termed nonresponders [31]. Some authorities, however, term a patient nonresponder after the failure of 6 injections [29].

However, other definition of non-responder includes no or minimal reduction in retinal thickness on OCT or no improvement in visual acuity. The study by DRCR. net defined less than 10% decrease in central subfield thickness on OCT and < 5 letter increase in visual acuity as no response to anti-VEGF treatment [21]. Options to treat such cases include other anti-VEGF agents, intravitreal triamcinolone, implantable steroid injection, macular laser, and targeted retinal photocoagulation (TRP) of peripheral capillary nonperfusion areas [30, 31, 33].

Center-involved diabetic macular edema is defined as retinal thickening involving the central subfield zone of the macula that is 1 mm in diameter [34]. The management of center-involved macular edema causing visual decline (visual acuity worse than 20/30) is relatively straightforward and such cases need treatment [28, 35]. The preferred therapy includes intravitreal anti-VEGF agents, steroids, steroid implants, or a combination of these. Cases with center-involved macular edema and good visual function pose a challenge to the treating Ophthalmologist. The dilemma in such cases is whether to start intravitreal therapy or to observe [30, 34]. Such cases have been reported to improve with good control of blood sugar levels alone [31]. The role of anti-VEGF agents in such cases is being explored [36]. These cases have to be monitored at regular intervals to detect deterioration in vision which is an indication to begin anti-VEGF therapy [31, 34].

Non-center involved diabetic macular edema is defined as a retinal thickening in the macula that does not involve the central subfield zone of diameter 1 mm [34]. Laser photocoagulation is still the standard of care for the treatment of cases with non-center involving macular edema [37]. For cases with macular edema with vitreomacular traction, induction of posterior vitreous detachment during pars plana vitrectomy with or without ILM peeling is the recommended choice for treatment [38–40].

#### **6. Biomarkers of DR on OCT**

Biomarkers are markers used externally to assess a medical state reliably and accurately [41]. Biomarkers may be physical, chemical, or biological. They are used to assess a physiological state, pathological process, or response to any pharmacological intervention [41]. Imaging biomarkers have the advantage of being non-invasive, reliable, and accurate. Several OCT-based biomarkers have been reported in DME which help in the management of the disease as well as in prognostication [42].

#### **6.1 Disorganization of the retinal inner layers (DRIL)**

Earlier studies showed a variable correlation between central retinal thickness measured on OCT and visual acuity achieved post-treatment of DME [43, 44]. A study by DRCR.net revealed that this correlation is modest. They also documented cases with a paradoxical decrease in visual acuity with a decrease in retinal thickening [45]. Further studies documented the role of OCT-based markers other than the central retinal thickness that affect visual acuity.

These include bridging retinal processes, the integrity of ELM and EZ, the reflectivity of cone outer segment tips, presence of hyperreflective foci, and subretinal fluid [46–49].

Long-standing cystoid macular edema with disturbance in ELM and EZ may suggest a poor visual outcome after treatment (**Figure 5**).

Sun and colleagues evaluated a novel marker in OCT, called disorganization of the retinal inner layers (DRIL), within the central 1 mm area of the fovea [50]. They studied the inner retinal layers in cases with existing DME or resolved DME. DRIL is 'defined as the horizontal extent in microns for which any boundaries between the ganglion cell–inner plexiform layer complex, inner nuclear layer, and outer plexiform layer could not be identified.' [50]. DRIL was found to have a substantial association with visual acuity. The presence of DRIL explained the paradoxical decrease in visual acuity in cases with resolved DME [50]. Later, Joltikov et al. reported the presence of DRIL in diabetics even before the presence of DR, DME, or PDR [51]. Further, Pelosini et al. proposed a theory to explain the negative correlation between retinal volume and visual acuity [52]. They suggested that the accumulation of fluid within the inner retinal layers causes the bipolar cells to stretch. Bipolar cells connect the photoreceptors to the ganglion cells. Fluid exceeding the limit of elasticity of these bipolar cell axons, may break the continuity of these axons and affect the transmission of signals between ganglion cells and photoreceptors. The irreversible destruction of bipolar cells provides a plausible explanation for cases with no improvement in visual acuity even after the resolution of DME [52]. In another study, the presence of retinal tissue between the cystic cavities in cases with DME was found to predict improvement in visual acuity after anti-VEGF therapy. These retinal tissues comprise of Müller and bipolar cells that transmit impulses between inner and outer retinal layers. The absence of these retinal bridging tissues at baseline explains the foveal thinning after the resolution of edema [53].
