Comparative Study between Intravitreal Ranibizumab Injection and Subthreshold Micropulsed Diode Laser (DSM) in Management of Diabetic Macular Edema

Document Type : Original Article

Authors

1 Opthalmology department, faculty of medicine, Al Azhar university,Cairo,Egypt

2 Ophthalmology department, faculty of medicine, Al Azhar university, Cairo

Abstract

ABSTRACT
Background: Diabetic retinopathy is a widely prevalent squeal of diabetes mellitus with multiple risk factors and predisposing conditions,
Objective: To compare efficacy of intravitreal injection of ranibizumab versus subthreshold micropulsed diode laser (DSM) in treatment of diabetic macular edema.
Patients and Methods: This was a prospective, comparative and interventional study that was performed in the departments of ophthalmology at Al Azhar University hospitals between September 2018 and January 2021.
Results: Improvement in the visual acuity (VA) was seen in both groups, The median VA in group (I) had changed from (0.21) to (0.25) and In group (II) it had changed from (0.26) to (0.30). Reduction in the central macular thickness (CMT) was seen in both groups. The median CMT in group (I) had decreased from 347.84 m to 322.48 μm and in group (II), it decreased from 333.04 m to 305.48 m. No intraoperative complication occurred during injection, the only postoperative complication that occurred were subconjunctival hemorrhage which occurred in 4 eyes (16%) and floaters which occurred in 7 eyes (28%). No complication occurred during laser treatment.
Conclusion: In our study we found that both intravitreal ranibizumab injection and 810 subthreshold micropulse laser when used with adequate power settings in the 5% duty cycle modes were effective in maintaining or improving visual acuity and improving central foveal thickness in the 50 eyes studied.

Keywords


INTRODUCTION

Diabetic Retinopathy (DR) is a common and specific microvascular complications of diabetes which affects 17–54% of people with diabetes aged between 49–60 years.1,2

Diabetic macular edoema (DME) is one of the most common causes of visual loss in today's society. 3 It affects about 10% of diabetic individuals and 29% of those who have had the condition for more than 20 years.4

Laser photocoagulation (focal, grid, or diode micropulse) vitrectomy, intravitreal injection of Ranibizumab, or triamcinolone acetonide have all been studied as treatment options for DME. The efficacy of these treatments was examined using best-corrected visual acuity (BCVA) and macular thickness determined using optical coherence tomography (OCT). There is a link between (BCVA) and OCT-measured macular thickness, however the relevance of this link is debatable.5

Although conventional laser therapy is the gold standard for slowing the progression of DR, it can sometimes exacerbate macular edoema. Short-pulse laser therapy, which was recently developed, is faster, produces less heat, and is less painful to the

 

 

 

 

 

 

eyes than traditional laser therapy. Furthermore, compared to the traditional pulse duration, short-pulse laser (diode laser micropulse) treatment causes less inflammation, fewer up-regulation of inflammatory cytokines following pan-retinal photocoagulation (PRP), and less macular thickening in patients with DR.6

Intravitreal injections of Ranibizumab have recently been shown to be useful in the treatment of DME, with good visual results. 7

Ranibizumab (Lucentis®, Genentech, Inc., South San Francisco, CA) is a recombinant humanized monoclonal antibody Fab fragment that binds and inhibits all isoforms of VEGF-A. Two pilot studies of ranibizumab demonstrated some efficacy in the treatment of DME. 8

According to the above data, this study was designed to compare the efficacy of intravitreal ranibizumab injections versus subthreshold micropulsed diode laser (DSM) in  treatment of diabetic macular edoema.

 

 

PATIENTS AND METHODS

Patient selection:

This was a prospective, comparative and interventional study that was performed in the departments of ophthalmology at Al Azhar University hospitals between September 2018 and January 2021.

This study involved 50 eyes with DME divided into two groups:Group I: Included (25) eyes where intravitreal ranibizumab injection (0.3mg in 0.05ml) was done. Group II:     Included (25) eyes where subthreshold micropulsed diode laser was done.

Inclusion criteria: Type 2 diabetis mellitus, HA1C less than 10, Phakic patient and pseudophakic patient with past history of cataract extraction since three months or more, Clinically and angiographically diagnosed as DME and have central foveal thickness (CFT) between 250um and 400um on  (OCT).

Exclusion criteria: Prior photocoagulation of the macula with a focal/grid laser, Previous ocular surgery, Pan-retinal photocoagulation, Diabetic papillopathy, Active intraocular inflammation, Drop of vision as a result of other causes, Past history of vitrectomy and past history of intravitreal injection of anti-VEGF or steroid.

Methods:

History TakingAge and sex, Diabetic history regarding the type and duration of diabetes.

Ophthalmological examination:

All the patients received complete ophthalmic examinations before injection and laser, incorporating distance (BCVA): Regarding the measurement of visual acuity before and after therapy by using decimal Snellen charts, Intraocular pressure (IOP) by usingGoldmann applanation tonometer. Examination of the anterior segment,including the presence of lens opacities and the presence of iris neovascularization, Posterior segment examination: using slit lamp and + 90 D and + 78 D lens to asses the vitreous for opacities or epiretinal membrane, The retina was assessed for any coexistent diseases, the stage of DR and the periphery of the retina was also examined carefully for any retinal holes or tears using the indirect ophthalmoscope.

Informed consent: The patients signed consent for intervention including: Advantages, disadvantages and the risks of possible complications.

For intravitreal injection: All injections were done in the operating rooms under complete aseptic conditions, the conjunctiva was anaesthetized first by 0.4 % oxybuprocaine eye drops (Benox).

Standard sterilization using Povidone-Iodine 5% (Betadine®) by lid swabbing and instillation in the conjunctival sac was done. A sharp 30-G needle was used to inject a volume of 0.05ml containing 0.3mg of Ranibizumab (Lucentis®, Genentech, Inc., South San Francisco, CA) into the vitreous cavity in phakic and pseudophakic eyes at distances of 4 and 3.5 mm from the limbus respectively.

Aqueous humour was extracted as needed by paracentesis incision with a 30-G needle to avoid a rise in IOP.

For subthreshold micropulsed diode laser: Using the micropulse mode of the Iridex OcuLightTM Tri-Mode SLx 810 nm laser. In this research, 23 patients were treated with a (DSM)

810 nm by using a (× 1.05 magnification) Mainster focal grid contact lens The following treatment parameters were employed in all cases: Exposure time was 200 milliseconds, the spot size was 200 millimetres, the power was 400 milliwatts, and the duty cycle was 5%. Micropulse technology's on and off cycles are referred to as duty cycle. The laser was turned on for 10 milliseconds and then turned off for 190 milliseconds. To guarantee that the surrounding tissues have enough time to cool before the inner retina is exposed to heat injury.

All patients will be followed up as follows:  Visual acuity testing (BCVA),  Full slit lamp examination, Intraocular pressure (IOP) measurement, Fundus examination, (OCT), at one month, three months and six months intervals.

Statistical analysis

The statistical program for social sciences, version 23.0, was used to analyse the data (SPSS Inc., Chicago, Illinois, USA). When the distribution was parametric, the quantitative values were provided as mean, standard deviation, and ranges (normal).

The tests that were carried out were as follows: Independent-samples The t-test of significance was used to compare two means., and the Mann Whitney U test was used for two-group comparisons in non-parametric data. When the predicted count in any cell was less than 5, the Chi-square test and Fisher's exact test were used instead of the Chi-square test to compare groups with qualitative data.


RESULTS

Baseline characteristics

Group I
(n=25)

Group II
(n=23)

Test value

p-value

Age (years)

 

 

 

 

Range

47-68

45-65

t=1.812

0.076

Mean±SD

57.97±7.12

54.90±4.07

Sex

 

 

 

 

Male

11/25 (44%)

8/23 (34.8%)

x2=0.127

0.721

Female

14/25 (56%)

15/23 (65.2%)

Duration of disease (years)

 

 

 

 

Range

7–19

5–24

t=1.719

0.092

Mean±SD

14.24±3.05

16.07±4.27

HbA1C

 

 

 

 

Range

6.4-9.2

6.7–9.5

t=1.286

0.205

Mean±SD

7.5±1.3

6.9±1.9

IOP

 

 

 

 

Range

12–18

13–21

t=1.000

0.323

Mean±SD

15.26±3.15

16.27±3.84

Laterality

 

 

 

 

Unilateral

25/25 (100%)

21/23 (91.3%)

FE

0.471

Bilateral

0/25 (0%)

2/23 (8.7%)

State of crystalline lens

 

 

 

 

Phakic

18/25 (72%)

16/25 (64%)

x2=0.092

0.762

Pseudophakic.

7/25 (28%)

9/25 (36%)

Table 1: Comparison between studied groups according to baseline characteristics.

shows There is no statistically significant difference between the two groups., with p-value (p >0.05 NS).

Regarding BCVA:

 At the base line: the mean BCVA in group (I) was 0.21 with SD 0.13 ranging from 0.1 to 0.5

 whereas in group (II) it was 0.26 with SD 0.12 ranging from 0.1 to 0.5 without statistically

significant difference between the two groups ( p-value 0.198).

 At one month : the mean BCVA was 0.29 with SD ±0.17 ranging from 0.1 to 0.7 in group (I) with statistically significant difference from the base line and was 0.28 with SD ±0.16 ranging from 0.1 to 0.6 in group (II) with  no statistically significant difference from the base line.

     At three months : the mean BCVA was 0.28 with SD ±0.17 ranging from 0.1 to 0.6 in group (I) with statistically significant difference from the base line and was 0.29 with SD ±0.15 ranging from 0.08 to 0.6 in group (II) with no statistically significant difference from the base line.

  At six months : the mean BCVA was 0.25 with SD ±0.17 ranging from 0.083 to 0.6 in group (I) with statistically non significant difference from the base line and was 0.30 with SD ±0.14 ranging from 0.08 to 0.6 in group (II) with no statistically significant difference from the base line and between the two groups, there is no statistically significant difference as shown in (table 2).

BCVA

Group I
(n=25)

Group II
(n=25)

Test value

p-value

Baseline

 

 

 

 

Mean±SD

0.21±0.13

0.26±0.12

U=-1.305

0.198

Range

0.1–0.5

0.1–0.5

After 1 month

 

 

 

 

Mean±SD

0.29±0.17#

0.28±0.14

U=-0.227

0.821

Range

0.1–0.7

0.1–0.6

After 3 months

 

 

 

 

Mean±SD

0.28±0.17#

0.29±0.15

U=0.221

0.826

Range

0.1–0.6

0.08–0.6

After 6 months

 

 

 

 

Mean±SD

0.25±0.17

0.30±0.14

U=-1.143

0.259

Range

0.083–0.6

0.08–0.6

Table 2 : Comparison between studied groups according to BCVA.This table shows statistically significant difference between baseline with after 1m and after 3m in group I, with p-value (p<0.05); while in group II there is no statistically significant difference with baseline, with p-value (p>0.05 NS) and statistically non significant difference between both groups, with p-value (p >0.05 NS).

p-value >0.05 NS; *p-value

Changes in the BCVA between baseline and over the period :

In group (I) the mean difference in BCVA (±SD) was 0.08(±0.18) with change (38.10%) , 0.07(±0.13) with change (33.33%), and 0.04 (±0.07)with change (19.05%) at 1, 3 , and 6, months, respectively. While In group (II), the mean difference in BCVA (±SD) was 0.02(±0.05) with change(7.70%) , 0.03(±0.06) with change (11.54%), and 0.04 (±0.09)with change (15.38%) at 1, 3 , and 6, months, respectively and no statistically significant difference exists between the two groups as shown in (table 3).

AT the end of the study, there were 13 eyes (52%) have the same BCVA , 10 eyes (40%) have increase in BCVA  and 2 eyes (8%) have decrease in BCVA in group (I) , while there were 8 eyes (32%) have the same BCVA , 14 eyes (56%) have increase in BCVA  and 3 eyes (12%) have decrease in BCVA in group (II) and no statistically significant difference exists between the two groups as shown in (table 4).

 

 

Change in BCVA  between baseline and over the period 

Group I
(n=25)

Group II
(n=25)

Test value

p-value

After 1 month

 

 

 

 

Mean Diff.±SD

0.08±0.18

0.02±0.05

U=1.606

0.115

Change%

38.10%

7.70%

After 3 months

 

 

 

 

Mean Diff.±SD

0.07±0.13

0.03±0.06

U=1.397

0.169

Change%

33.33%

11.54%

After 6 months

 

 

 

 

Mean Diff.±SD

0.04±0.07

0.04±0.09

U=0.044

0.965

Change%

19.05%

15.38%

Table 3: Comparison between studied groups according to change in BCVA between baseline and over the period (there was no statistically significant difference between both groups, with p-value (p >0.05 NS).

BCVA after 6months

Group I
(n=25)

Group II
(n=25)

Test value

p-value

The same

13 (52%)

8 (32%)

2.057

0.358

Increase

10 (40%)

14 (56%)

Decrease

2 (8%)

3 (12%)

Table 4: Comparison between studied groups according to BCVA after 6months.which shows that between the two groups, there was no statistically significant difference. with p-value (p>0.05 NS).

Regarding CMT:

At the base line: the mean CMT was 347.84 µm with SD ±34.95 µm ranging from 269 to 390 µm  in group (I) andwas 333.04  µm with SD ±37.85 µm ranging from264 to 391 µm in group (II) with no statistically significant difference between both

groups with ( p-value 0.157).

At one month : the the mean CMT was 292.48 µm with SD ±39.63 µm ranging from 210 to 358 µm in group (I)with statistically significant difference from the base  line and was 324.44 µm with SD ±39.45 µm ranging from 260 to 390 µm in group  (II) with no statistically significant difference from the base line and also between the

two groups, there is a statistically significant difference. with ( p-value 0.006).

 At three months : the the mean CMT was 310.96 µm with SD ±46.18 µm ranging from 210 to 372µm in group (I) with statistically significant difference from the base line and was 308.24 µm with SD ±44.20 µm ranging from 245 to 395 µm in group (II) with statistically significant difference from the base line and between the two groups, there was no statistically significant difference.

At six months : the the mean CMT was 322.48 µm with SD ±53.36 µm ranging from 213 to 397µm in group (I) with statistically significant difference from the base line and was 305.48 µm with SD ±48.70 µm ranging from 240 to 400 µm in group (II) with statistically significant difference from the base line and between the two groups, there was no statistically significant difference as shown in (table 5).

CMT

Group I
(n=25)

Group II
(n=25)

Test value

p-value

Baseline

 

 

 

 

Mean±SD

347.84±34.95

333.04±37.85

t=1.436

0.157

Range

269–390

264–391

After 1 month

 

 

 

 

Mean±SD

292.48±39.63#

324.44±39.45

t=2.858

0.006*

Range

210–358

260–390

After 3 months

 

 

 

 

Mean±SD

310.96±46.18#

308.24±44.20#

t=0.213

0.832

Range

210–372

245–395

After 6 months

 

 

 

 

Mean±SD

322.48±53.36#

305.48±48.70#

t=1.177

0.245

Range

213–397

240–400

Table 5: Comparison between studied groups according to CMT which shows statistically significant difference between groups according to CMT after 1 months, with p-value (p

Reduction in CMT:

 The mean reduction in CMT as compared to baseline (±SD) was -55.36(±45.40), -36.88(±32.09), and -25.36(±23.08) µm at 1, 3, and 6 months respectively in group (I) and -8.6 (±7.05), -24.8 (±21.58), and  -27.56 (±25.08) at 1, 3, and 6 months, respectively in group (II) as shown (table 6).

Percentage of reduction in CMT:

The mean percentage of reduction in CMT was -15.92%, -10.60%, and -7.29% at 1,3,6 Months respectively in group (I) and was -2.58%, -7.45%, -8.28% at 1,3 and 6 Months respectively in group (II) as shown (table 6).

There was no statistically significant difference in the percentage of reduction of CST between the two studied groups except at 1 months as shown in table (6).

Change in CMT between baseline and over the period

Group I
(n=25)

Group II
(n=25)

Test value

p-value

After 1 month

 

 

 

 

Mean Diff.±SD

-55.36±45.40

-8.6±7.05

U=5.089

<0.001**

Change%

-15.92%

-2.58%

After 3 months

 

 

 

 

Mean Diff.±SD

-36.88±32.09

-24.8±21.58

U=1.562

0.125

Change%

-10.60%

-7.45%

After 6 months

 

 

 

 

Mean Diff.±SD

-25.36±23.08

-27.56±25.08

U=0.323

0.745

Change%

-7.29%

-8.28%

Table 6: Comparison between studied groups according to change in CMT between baseline and over the period which shows statistically significant difference after 1months, with p-value (p

 

CASE PRESENTATION

 

Fig. 1: Case 5 group I (upper left image) OCT baseline 386µm, (upper right image) OCT 1st month 315 µm, (lower left image) OCT 3rd month 339 µm, (lower right image) OCT 6th month 372 µm.

 

Fig. 2: case 1 group II (upper left image) OCT baseline 350µm, (upper right image) OCT 1st month 338 µm, (lower left image) OCT 3rd month 320 µm, (lower right image) OCT 6th month 304 µm


 

 

 

 

DISCUSSION

In individuals aged 20 to 74 years, DR is one of the primary causes of vision loss. 3

DME is a leading cause of vision loss in persons with diabetes mellitus.9

Available therapies include laserphotocoagulation either conventional or micropulse, corticosteroids, and anti-VEGFdrugs.10

Laser photocoagulation (LPC) was the sole viable treatment for DME that preserved true visual acuity for many years.11

However, because this type of therapy involves the death of photoreceptors, It had a significant negative impact similar scotomas in the visual field and the risk of subsequent choroidal neovascularization. As a result, laser photocoagulation (LPC) is only utilised in the treatment of DME infrequently nowadays, mainly when other therapeutic options are unavailable or inappropriate. For many years, subthreshould micropulse laser therapy (SMPLT) has been utilised to treat DME as a non-damaging retinal therapy.12

In our study we also use 5% DC comparative with Luttrull et al 13. that use 10-15 % DC the frequency of laser induced retinal damage was 8% (7/84) eyes while using the 10-15% duty cycle and a 0% (0/164) eyes while using the 5% duty cycle. Using a 5% duty cycle showed no detectable damage.

In our research we included 50 eyes with clinically significant macular edema divided into 2 groups of 25 patients in group (I) and 23 patients in group (II). The first group (group I) was treated with intravitreal ranbizumab injection while the second group (group II) was treated with subthreshould micropulse diode laser.

The improvement in the anatomical (CMT) and functional (BCVA) outcomes in the present study may be attributed to patients’ selection criteria including exclusion of ischemic DME cases and absent history of previous laser or injection before enrollment.

Regarding VA and CMT:

The mean BCVA improved from 0.21 to 0.25 After 6 mo in group (I) and from 0.26 To 0.30 in group (II) With no statistically significant difference between both groups.

The mean CMT improved from 347.84 µm to 322.48 µm After 6 mo in Group (I) And from 333.04 µm to 305.48 µm In group (II) Without statistically significant difference between both groups.

After 1st month of treatment the mean central macular thickness improved from 347.84 to 292.48 µm in Group (I) And from 333.04 µm to 324.44 µm In group (II) With statistically significant difference between both groups with (p-value 0.006).

Nakamura et al 14. used subthreshold micropulse diode laser photocoagulation in 28 eyes with DME to examine the Functional and morphological changes of macula and observed a significant decrease in CMT from 481 m to 388 m after 3 months and a significant improvement in BCVA from 0.47 logMAR to 0.4 logMAR. There was no discernible difference in retinal sensitivity.

 

Mansouri et al 15. studied 63 eyes with DME who were treated with SMPLT. A comparison of the outcomes (BCVA, CMT) in groups 1 and 2 with macular edema (ME) 400m (33 eyes) and ME > 400m (30 eyes) respectively, Following a six-month follow-up Group 1 demonstrated a significant improvement in BCVA of 0.2 logMAR on average, as well as a significant reduction in CMT of 55m on average. No significant change in BCVA or CMT in Group 2.

 According to Dorin16. Despite the large number of publications published in the literature, micropulse photocoagulation for the treatment of macular retinal vascular disease is not commonly used. The practitioner's inability to see the intraoperative laser tissue reaction could be one cause for this resistance. This automatically establishes a psychological barrier to applying an invisible spot to the retina and being unable to identify the location of the laser spots on follow-up.

Elman et al 17 .  reported that Previous major clinical trials after IVI of Anti-VEGFs have improvements in BCVA from baseline and associated with reductions in the CMT from baseline.

In center-involving DME, anti-VEGF drugs are considered the first line of treatment; nevertheless, all significant clinical trials have indicated that only 33–45 percent of DME patients on anti-VEGF medications demonstrate three lines or more of visual improvement.18,19,20

CONCLUSION

In the 50 eyes tested, both intravitreal ranibizumab injection and the 810 subthreshold micropulse diode laser were effective in preserving or increasing visual acuity and improving central foveal thickness when utilised with sufficient power settings in the 5 percent duty cycle modes. The DSM appears to be a new non-invasive and successful method of treating DME. Treatment with a subthreshold micropulse laser is inexpensive, safe, and painless for the patient; yet, even after a single ranibizumab injection and one session of subthreshold micropulse laser the visual result and CMT thickness were still better than before treatment. After 6 months of follow-up.

  1. REFERENCES

    1. Lim LS, Liew G, Cheung N, Mitchell P and Wong TY. Mixed messages on systemic therapies for diabetic retinopathy. The Lancet. 2010; 376(9751):1461.
    2. Macky TA, Khater N, Al-Zamil MA, El Fishawy H and Soliman MM. Epidemiology of diabetic retinopathy in Egypt: a hospital-based study. Ophthalmic research, 2011;45(2):73-8.
    3. Lee R, Wong TY and Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis (Lond). 2015;2:17.
    4. Jung YH and Lee Y. Efficacy of vitrectomy combined with an intraoperative dexamethasone implant in refractory diabetic macular edema. Acta Diabetol. 2019;56(6):691 6. 
    5. Otani T, Yamaguchi Y and Kishi S.  Correlation between visual acuity and foveal microstructural changes in diabetic macular edema. RETINA. 2010; 30:774–80.
    6. Takamura Y, Shimura M, Katome T, Someya H and Sugimoto M et al. Effect of intravitreal triamcinolone acetonide injection at the end of vitrectomy for vitreous haemorrhage related to proliferative diabetic retinopathy. British Journal of Ophthalmology. 2018; 102(10):1351-7.
    7. Moshfeghi AA, Rosenfeld PJ, Flynn Jr HW, Schwartz SG and Davis JL et al. Endophthalmitis after intravitreal anti–vascular endothelial growth factor antagonists: A Six-Year Experience at a University Referral Center. Retina. 2011; 31(4):662-8.
    8. Nguyen QD, Tatlipinar S and Shah SM. Vascular endothelial growth factor is a critical stimulus for diabetic macular edema. Am J Ophthalmol. 2006;142(6):961-9.
    9. Varma R, Bressler NM, Doan QV, Gleeson M and Danese M et al. Prevalence of and risk factors for diabetic macular edema in the United States. JAMA Ophthalmol. 2015; 132(11).
    10. Zur D and Loewenstein A. Combination therapy for diabetic macular edema. J Ophthalmologica 2012; 2012:484612.‎
    11. Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema.Early treatment diabetic retinopathy study report number 1. Arch. Ophthalmol. 1985; 103:1796–806.
    12. Berger AR and Boniuk I. Bilateral subretinal neovascularization after focal argon laser photocoagulation for diabetic macular edema. Am. J. Ophthalmol. 1989; 108: 88–90.
    13. Luttrull J, Musch D and Mainster M. Subthreshold diode micropulse photocoagulation for the treatment of clinically significant diabetic macular oedema. Br J Ophthalmol. 2005;89(1):74-80.
    14. Nakamura Y, Mitamura Y, Ogata K, Arai M and Takatsuna Y et al. Functional and morphological changes of macula after subthreshold micropulse diode laser photocoagulation for diabetic macular oedema. Eye. 2010; 24:784–8.
    15. Mansouri A, Sampat KM, Malik KJ, Steiner JN and Glaser BM. micropulse laser in the treatment of diabetic macular edema is influenced by pre-treatment central foveal Medscape. Efficacy of subthreshold thickness. Eye. 2014; 28:1418–24.
    16. Dorin G. Subthreshold and micropulse diode laser photocoagulation. Semin Ophthalmol. 2003; 18(3):147-53.
    17. Elman MJ, Ayala A, Bressler NM, Browning D and Flaxel CJ et al. Intravitreal ranibizumab for diabetic macular edema with prompt versus deferred laser treatment: 5-year randomized trial results. Ophthalmology. 2015; 122:375–81.
    18. Schmidt-Erfurth U, Garcia-Arumi J and Bandello F. Guidelines for the Management of Diabetic Macular Edema by the European Society of Retina Specialists (EURETINA). Ophthalmologica. 2017; 237 (4) : 185‐222.
    19. Urias EA, Urias GA, Monickaraj F, McGuire P and  Das A. Novel therapeutic targets in diabetic macular edema: beyond VEGF. Vision Research. 2017; 139:221–7.
    20.  Browning DJ, Stewart MW and Lee C. Diabetic macular edema: evidence-based management. Indian Journal of Ophthalmology. 2018; 66(12):1736–50.