2.*Dexamethasone intravitreal implant (DEX implant; Ozurdex; Allergan, Inc., Irvine, CA, USA):*

OZURDEX® (dexamethasone intravitreal implant) 0.7 mg is a biodegradable implant injected into the vitreous to treat diabetic macular edema. Ozurdex was approved for treatment of diabetic macular edema in late 2014 [88].

In the MEAD trials, a great percentage of patients achieved a visual gain of ≥15-letters from baseline with improved central retinal thickness (CRT) by OCT [89].

In the PLACID trial, comparing Ozurdex monotherapy to laser monotherapy, a higher percentage of patients treated with dexamethasone achieved at least 10 letters better improvement as compared to those treated with laser monotherapy [90].

Dexamethasone has lower lipophilic properties than other corticosteroids, with consequent lower binding affinity to the trabecular meshwork. Thus, dexamethasone implants are associated with lower risks of glaucoma and cataract as compared to other steroids used for treatment of DME [91].

3.*Fluocinolone acetonide (FA) intravitreal implant:* It is a nonerodible implant approved for the treatment of diabetic macular edema (DME). The injectable intravitreal implant releases fluocinolone acetonide at a rate of about 0.2 μg/day for a duration of about 36 months [92].

The Fluocinolone Acetonide in Diabetic Macular Edema (FAME) trials showed that, in patients with center-involved DME previously treated with laser photocoagulation, FA intravitreal implant was superior as compared to sham injection in improving VA more than 15 letters at 24 months and at 36 months [93, 94].

The two commonest side effects of the drug were cataract (86%) and nearly all of them required cataract extraction and intraocular pressure (IOP) elevation (37%). Raised IOP was mostly treated with antiglaucoma medications, with <5% of eyes requiring glaucoma surgery [95].

### **3.4 Guidelines for treatment**

#### *3.4.1 Subclinical DME (SCME)*

Subclinical macular edema is defined as thickening of the center of the macula identified by OCT but not detected on clinical examination [96].

Monitoring the progression of subclinical DME (SCME), the Diabetic Retinopathy Clinical Research Network (DRCR.net) showed that about a quarter to half of eyes would progress to clinically significant macular edema (CSME) within 2 years, and are 12 times more prone to the development of CSME as compared to eyes without SCME, particularly if there is increased retinal thickening within the outer ring of the ETDRS grid. This is an important indicator that SCME is an important biomarker for development of center-involving diabetic macular edema (CIME) and this was also found to be commoner in female patients [96, 97].

On the other hand, there is a controversy as to whether the level of HBA1c is related to the progression of SCME to CIME or not [97, 98].

The management of subclinical DME is mainly the prevention of its progression to CSME. Scheduled follow-up visits, with OCT performed together with glycemic control, and strict control of the other systemic risk factors mentioned above are recommended [43].

#### *3.4.2 Treatment recommendations for non-center-involving DME (non-CIME)*

Noncenter-involving DME represents a precursor to the development of centerinvolved clinically significant ME. If not treated, it was found that 38% of eyes will gain at least 50 μm in CMT over 2 years [96].

Laser treatment is recommended for this subset of DME by most authors.

A modified ETDRS (mETDRS) laser treatment was recommended by the Diabetic Retinopathy Clinical Research (DRCR) net study and was based on treating areas of thickened macula (as shown by OCT imaging) and areas of nonperfusion (detected by OCTA and FA) and leaking microaneurysm (detected by FA) with less intense and smaller burns than in the original ETDRS treatment. Results of this study showed a gain of 15 letters in 25% of the patients [99]. Complications of this technique however include central scotomata and loss of central vision caused by progressive enlargement of the laser scars [100].

Focal laser is another option adopted where focal laser is applied to all leaking microaneurysms 500–3000 μm away from the fovea. Results of this line of treatment showed improvement of visual acuity (VA) in 21% of eyes and stabilization of baseline VA in 61% of eyes [101].

One study, however, recommended either observation until involvement of the center of the macula or alternatively initiating anti-VEGF therapy or focal laser photocoagulation [58].

#### *3.4.3 Treatment recommendations for center-involved DME*

Most of the guidelines for center-involved DME recommend anti-VEGF intravitreal injections as the first line of treatment, unless contraindicated as in cases of a recent cardiovascular or cerebrovascular event within the previous 3–6 months, breast feeding and pregnancy [43, 58, 102–104].

Anti-VEGF injection is currently loaded by following one of the following protocols:

1.Treat and extend protocol (T&E)

With Ranibizumab, after the initial 3-monthly loading doses, the patient is switched to a T&E regimen if there were no signs of disease activity as indicated by

#### *Diabetic Macular Edema, Clinicopathologic and Keys for Management DOI: http://dx.doi.org/10.5772/intechopen.112974*

OCT images and the best-corrected visual acuity (BCVA) was either the same or better as compared to that at the last visit. In this regimen, follow-up visits were extended by a 4-week interval so long as there was no disease activity for a maximum duration of 24 weeks. If signs of activity appeared (recurrence of intraretinal or subretinal fluid or CRT more than 300 μm), a new injection is given and the follow-up duration is shortened by 4 weeks to a minimum interval of 4 weeks between the visits [105].

For aflibercept, the recommended regimen is five loading monthly injections followed by one injection every 2 months for the first year. Once stabilized, the duration could be extended as above [106].

#### 2.Pro re nata (PRN) regimen

In this regimen, the patient received the usual loading dose followed by monthly injections until there was no disease activity by OCT and the best-corrected visual acuity (BCVA) was either stabilized or improved as compared to the previous visit. Patients then were scheduled for regular follow-ups every 1–3 months and received no further injections, unless recurrence of disease activity was noted on OCT [105].

In a recent meta-analysis comparing T&E and PRN protocols for DME, there was no clear advantage in reducing the number of injections between the two groups; however, there were, in the T&E regimen, limited gains in visual and anatomical outcomes. The T&E regimen also allows for fewer patient visits, thereby reducing treatment burden [107].

All guidelines agree that nonresponders to anti-VEGF treatment (after the first 3–6 injections) should be switched to another anti-VEGF or to steroids. The definition of (nonresponders), however, differed in various guidelines. In one *post hoc* analysis exploring the relationship between early retinal anatomical (CMT on OCT imaging) and functional (visual acuity) responses (after 12 weeks) and long-term same anatomical and visual outcomes (weeks 52 and 156) in eyes treated with ranibizumab plus prompt or deferred laser (in Protocol I study), a conclusion was set that a reduction of less than 20% in central retinal thickness or less than 5-letter improvement in VA is to be categorized as (nonresponder) [108, 109].

The criteria of (nonresponders) in the American Delphi Panel guidelines were however different, they considered that failure to achieve a visual acuity of 20/40 or better or failure to achieve a reduction of at least 50% of excess macular thickness on OCT after the initial 3–6-month loading dose is to be considered nonresponsiveness [102].

The Spanish Delphi panel also had different criteria for nonresponders. They considered a less than 10% reduction in CRT or a less than 5 letter improvement in VA a criterion for nonresponsiveness [103].

For nonresponders, a switch to another anti-VEGF is recommended if the patient is phakic, to lower the incidence of cataract and glaucoma development. If the DME remained without responding, a switch to steroids is recommended [43]. Dexamethasone implant is the first choice in the steroid group followed by fluocinolone acetonide intravitreal implant in case of steroid nonresponsiveness [110].

A steroid dexamethasone implant (DEX implant; Ozurdex) can be used as a first-line treatment if anti-VEGF is contraindicated, in the above-mentioned conditions, or if there is poor compliance with the anti-VEGF regimen (because of the frequent visits or social or economic issues), and may also be recommended in patients who are vitrectomized, pseudophakic (because cataract development is not now an issue) or with chronic DME [111, 112]. IOP measurement and monitoring

the development of cataract should be regularly checked in patients in whom a steroid implant is used [112].

One study suggested that DME associated with an increased inflammatory response, especially the cystoid and the sensory neural detachment types, as mentioned above, may resolve better with steroids as compared to anti-VEGFs [14].

#### *3.4.3.1 Management of center-involved DME in pregnancy*

Early Diabetic Treatment Retinopathy Study (ETDRS) recommended the use of grid or focal laser for clinically significant macular edema (CSME) in general. This was, however, before the era of anti-VEGFs and steroids [113]. ETDRS guidelines were, however, not specific for pregnancy, but two studies [114, 115] supported the use of macular laser at the earliest to prevent irreversible damage. Another study [116] recommended the use of micropulse laser for foveal involvement, as it is safer.

Anti-VEGF and triamcinolone are better to be avoided during pregnancy for fear of teratogenicity in cases where there is refractory DME. Since steroids were proven to be more effective in improving VA after 4 months of follow-up as compared to laser as proved by the Diabetic Retinopathy Clinical Research (DRCR) Network [117], it is therefore recommended to use dexamethasone implant for DME that developed before pregnancy [43].

Observation is a reasonable management option for pregnant patients with mild diabetic macular edema (DME) or DME developing during pregnancy since the edema may well resolve after delivery [43, 118].

The American Academy of Ophthalmology Preferred Practice Patterns recommend that pregnant diabetic patients undergo dilated fundus examination in the first trimester, with subsequent follow-up determined by the severity of retinopathy, as per every 3–6 months for moderate nonproliferative diabetic retinopathy (NPDR) and those with severe NPDR or worse should be examined every 1–3 months. *Postpartum* follow-up should continue during the first year [117].
