Current status in diabetic macular edema treatments

MHANet: A hybrid attention mechanism for retinal diseases classification

With the increase of patients with retinopathy, retinopathy recognition has become a research hotspot. In this article, we describe the etiology and symptoms of three kinds of retinal diseases, including drusen(DRUSEN), choroidal neovascularization(CNV) and diabetic macular edema(DME). In addition, we also propose a hybrid attention mechanism to classify and recognize different types of retinopathy images. In particular, the hybrid attention mechanism proposed in this paper includes parallel spatial attention mechanism and channel attention mechanism. It can extract the key features in the channel dimension and spatial dimension of retinopathy images, and reduce the negative impact of background information on classification results. The experimental results show that the hybrid attention mechanism proposed in this paper can better assist the network to focus on extracting thr fetures of the retinopathy area and enhance the adaptability to the differences of different data sets. Finally, the hybrid attention mechanism achieved 96.5% and 99.76% classification accuracy on two public OCT data sets of retinopathy, respectively.

Long-term outcomes of intravitreal therapy for symptomatic diabetic macular oedema in a real-world setting in Switzerland

Objective

To assess the long-term visual outcomes in eyes with symptomatic diabetic macular oedema (DME) under intravitreal treatment (IVT) in a clinical routine setting.

Methods

Patients with newly diagnosed DME were included in this retrospective study if they had received at least three IVTs and a follow-up period ≥ 2 years. Due to altered treatment patterns since the approval of ranibizumab for DME in 2012, patients were subdivided according to their first IVT before 2013 (group 1) or thereafter (group 2). The primary outcome measure was the evolution of best-corrected visual acuity (BCVA) over time.

Results

Of 217 eyes (191 patients) with DME, 151 eyes (117 patients) fulfilled the inclusion criteria (63 eyes in the first period, 88 in the second period). Mean follow-up time was 7.9 ± 3.1 (group 1) and 4.1 ± 1.4 years (group 2; p < 0.001). Visual gains were similar in the first year (group 1: + 5.3 ± 15.5, group 2: + 7.3 ± 12.2 Early Treatment Diabetic Retinopathy Study (ETDRS) letters; p = 0.44), but not thereafter (after 2 years in group 1: + 4.4 ± 15.0, group 2: + 8.3 ± 13.0 ETDRS letters; p = 0.038). During the first year, group 1 patients received less clinical examinations (group 1: 6.6 ± 3.3, group 2: 7.5 ± 2.1; p = 0.007) and less injections (group 1: 3.6 ± 2.7, group 2: 6.1 ± 2.7; p < 0.001).

Conclusion

A greater visual gain, in response to more intensive treatment during the first year, was maintained for at least 5 years in group 2 subjects. Our data confirm that in a real-world setting, early intensive treatment results in satisfying long-term visual outcomes.

Evaluation of vitrectomy combined preoperative intravitreal ranibizumab and postoperative intravitreal triamcinolone acetonide for proliferative diabetic retinopathy

BACKGROUND

To explore the treatment efficacy of the combination of preoperative intravitreal ranibizumab (IVR) and postoperative intravitreal triamcinolone acetonide (IVTA) in patients undergoing pars plana vitrectomy (PPV) for proliferative diabetic retinopathy (PDR).

METHODS

A retrospective comparative study was performed on 128 eyes of 128 patients who had PDR and underwent PPV. Patients who received a single PPV were assigned to Group A. Those who received PPV with preoperative IVR were assigned to Group B. Patients in Group C underwent PPV combined preoperative IVR and postoperative IVTA. Intraoperative findings, changes in mean best-corrected visual acuity (BCVA) and postoperative adverse events, were retrospectively evaluated at 6-month follow-up.

RESULTS

The incidences of iatrogenic breaks, severe intraoperative bleeding, using long-term internal tamponade agents, recurrent vitreous hemorrhage (VH), and duration of surgery were statistically significantly less in Group B and Group C than in Group A. The postoperative BCVA was statistically significantly better in Groups B and Group C than in Group A, respectively, at 1 month after surgery. The mean 3-month postoperative visual acuity was better in Group C. The incidence of high intraocular pressure (IOP) was significantly higher in Group C at the first postoperative week. There were no statistically significant differences in the incidence of exudative retinal detachment and choroidal detachment among the three groups.

CONCLUSION

In patients undergoing PPV for PDR, preoperative IVR significantly reduced the occurrence of intraoperative and postoperative complications, and the combination of preoperative IVR and postoperative IVTA can better improve the postoperative visual outcome.

Protective effects of hydrogen gas in a rat model of branch retinal vein occlusion via decreasing VEGF-α expression

Background

Oxidative stress (OS) is an essential factor in the pathogenesis of branch retinal vein occlusion (BRVO). Studies have demonstrated the role of hydrogen gas in the regulation of OS. This study was designed to evaluate the efficacy of hydrogen gas on the BRVO rat model.

Methods

Twenty-four BRVO rats were randomly divided into two groups: the hydrogen gas (H) group (42% H₂, 21% O₂, 37% N₂) and the model (M) group (21% O₂, 79% N₂). Rats in the H group inhaled hydrogen gas for 8 h every day up to 30 d post-occlusion. Twelve age-matched healthy rats served as the control (C) group. Retinal function and morphology were detected at 1, 7, 14 and 30 d post-occlusion. Furthermore, the expression of vascular endothelial growth factor (VEGF-α) was detected by immunofluorescent staining.

Results

Full-field electroretinography (ffERG) revealed that the amplitude of the b-wave (dark-adaptation 3.0 response), the amplitude of the OPs2 wave and the light-adapted flicker response in the H group were all higher than those in the M group at 7 d post-occlusion (all p < 0.05). The reopen time of occlusive retinal vessels in the H group was 2.235 ± 1.128 d, which was shorter than that in the M group (4.234 ± 2.236 d, p < 0.05). The rats in the H group had a thinner IPL + GCL + NFL and an increased total retina compared with those in the M group at 3 d post-occlusion (p < 0.05), while the rats in the H group had a thicker INL, IPL + GCL + NFL and total retina compared with those at 7, 14 and 30 d post-occlusion (p < 0.05). Moreover, the flow velocity of ear vein blood was increased in the H group compared with that in the M group (p < 0.05). The expression of VEGF-α in the H group was dramatically decreased compared with that in the M group at 1, 7 and 14 d post-occlusion (p < 0.05), while the expression kept in similar level at 30 d post-occlusion (p > 0.05).

Conclusions

Our findings demonstrate that inhalation of hydrogen gas could alleviate retinal oedema, shorten reopen time and improve retinal function, and the potential mechanism might be related to a decrease in VEGF-α expression.

Optimized Deep Convolutional Neural Networks for Identification of Macular Diseases from Optical Coherence Tomography Images

Finetuning pre-trained deep neural networks (DNN) delicately designed for large-scale natural images may not be suitable for medical images due to the intrinsic difference between the datasets. We propose a strategy to modify DNNs, which improves their performance on retinal optical coherence tomography (OCT) images. Deep features of pre-trained DNN are high-level features of natural images. These features harm the training of transfer learning. Our strategy is to remove some deep convolutional layers of the state-of-the-art pre-trained networks: GoogLeNet, ResNet and DenseNet. We try to find the optimized deep neural networks on small-scale and large-scale OCT datasets, respectively, in our experiments. Results show that optimized deep neural networks not only reduce computational burden, but also improve classification accuracy.

Fully automated detection of retinal disorders by image-based deep learning

PURPOSE

With the aging population and the global diabetes epidemic, the prevalence of age-related macular degeneration (AMD) and diabetic macular edema (DME) diseases which are the leading causes of blindness is further increasing. Intravitreal injections with anti-vascular endothelial growth factor (anti-VEGF) medications are the standard of care for their indications. Optical coherence tomography (OCT), as a noninvasive imaging modality, plays a major part in guiding the administration of anti-VEGF therapy by providing detailed cross-sectional scans of the retina pathology. Fully automating OCT image detection can significantly decrease the tedious clinician labor and obtain a faithful pre-diagnosis from the analysis of the structural elements of the retina. Thereby, we explore the use of deep transfer learning method based on the visual geometry group 16 (VGG-16) network for classifying AMD and DME in OCT images accurately and automatically.

METHOD

A total of 207,130 retinal OCT images between 2013 and 2017 were selected from retrospective cohorts of 5319 adult patients from the Shiley Eye Institute of the University of California San Diego, the California Retinal Research Foundation, Medical Center Ophthalmology Associates, the Shanghai First People's Hospital, and the Beijing Tongren Eye Center, with 109,312 images (37,456 with choroidal neovascularization, 11,599 with diabetic macular edema, 8867 with drusen, and 51,390 normal) for the experiment. After images preprocessing, 1000 images (250 images from each category) from 633 patients were selected as validation dataset while the rest images from another 4686 patients were used as training dataset. We used deep transfer learning method to fine-tune the VGG-16 network pre-trained on the ImageNet dataset, and evaluated its performance on the validation dataset. Then, prediction accuracy, sensitivity, specificity, and receiver-operating characteristic (ROC) were calculated.

RESULTS

Experimental results proved that the proposed approach had manifested superior performance in retinal OCT images detection, which achieved a prediction accuracy of 98.6%, with a sensitivity of 97.8%, a specificity of 99.4%, and introduced an area under the ROC curve of 100%.

CONCLUSION

Deep transfer learning method based on the VGG-16 network shows significant effectiveness on classification of retinal OCT images with a relatively small dataset, which can provide assistant support for medical decision-making. Moreover, the performance of the proposed approach is comparable to that of human experts with significant clinical experience. Thereby, it will find promising applications in an automatic diagnosis and classification of common retinal diseases.

Effect of Intravitreal Bevacizumab with or without Macular Photocoagulation for Diabetic Macular Edema: A Meta-Analysis

INTRODUCTION

In this meta-analysis, we aimed to assess the possible benefits of macular photocoagulation (MPC) as an additional treatment with intravitreal bevacizumab (IVB) in patients with diabetic macular edema.

METHODS

The studies were identified from three databases: PubMed, Web of Science, and the Cochrane Library. The main outcome measures included change in best-corrected visual acuity (BCVA), differences in central macular thickness (CMT), and adverse events within the follow-up period. The results were pooled using weight mean difference with their corresponding 95% confidence intervals. A fixed or random effects model was employed, depending on the heterogeneity of the inclusion trials.

RESULTS

Finally, three randomized controlled trial and two high-quality retrospective studies were identified and included. Changes in CMT at 1, 3, and 6 months did not vary significantly between the IVB-alone group and the IVB with MPC group (P = 0.26, 0.06, and 0.65, respectively). Similarly, changes in BCVA at 1, 3, and 6 months also did not vary significantly between the two groups (P = 0.20, 0.91, and 0.70, respectively). Whereas substantial heterogeneity was detected in the CMT results among these studies, the sensitivity analyses showed Solaiman's study was probably the source of the heterogeneity. No publication bias was detected by funnel plots in this study.

CONCLUSION

Results of this meta-analysis showed that the treatments with IVB alone and combined IVB and MPC were similarly effective in improving BCVA and reducing CMT. However, more evidence is needed to evaluate their effects in the long-term periods.

Multicenter, Randomized Clinical Trial to Assess the Effectiveness of Intravitreal Injections of Bevacizumab, Triamcinolone, or Their Combination in the Treatment of Diabetic Macular Edema

BACKGROUND AND OBJECTIVE

To evaluate the efficacy of combined bevacizumab-triamcinolone intravitreal injection in the treatment of diabetic macular edema (DME) compared to monotherapy.

PATIENTS AND METHODS

At eight clinical sites, 111 patients with DME were randomly assigned to receive an intravitreal injection of bevacizumab (Avastin; Genentech, South San Francisco, CA), triamcinolone (Ophthalmos Pharmaceutical Industry, São Paulo-SP, Brazil), or their combination. The primary outcome was visual acuity (VA) at 6 months' follow-up.

RESULTS

The average number of injections was 3.2 in the bevacizumab group, 2.4 in the combined group, and 2.1 in the triamcinolone group. All groups presented with improvements in VA (P < .001); however, no differences between groups were observed (P = .436). Mean reduction in central retinal thickness was statistically different only between the triamcinolone and bevacizumab groups (P < .015).

CONCLUSION

Mono- or combination therapy was effective for DME treatment. No synergistic effects were observed; however, triamcinolone alone or a drug combination may reduce the number of injections required when compared to bevacizumab alone. [Ophthalmic Surg Lasers Imaging Retina. 2017;48:734-740.].

Clear, Aqueous Topical Drop of Triamcinolone Acetonide

The objective of this study was to develop a clear aqueous mixed nanomicellar formulation (NMF) of triamcinolone acetonide (TA) with a combination of nonionic surfactant hydrogenated castor oil 60 (HCO-60) and octoxynol-40 (Oc-40). In order to delineate the effects of drug-polymer interactions on entrapment efficiency (EE), loading efficiency (LE), and critical micellar concentration (CMC), a design of experiment (DOE) was performed to optimize the formulation. In this study, full-factorial design has been used with HCO-60 and OC-40 as independent variables. All formulations were prepared following solvent evaporation and film rehydration method, characterized with size, polydispersity, shape, morphology, EE, LE, and CMC. A specific blend of HCO-60 and Oc-40 at a particular wt% ratio (5:1.5) produced highest drug EE, LE, and smallest CMC (0.0216 wt%). Solubility of TA in NMF improved 20 times relative to normal aqueous solubility. Qualitative ¹H NMR studies confirmed the absence of free drug in the outer aqueous NMF medium. Moreover, TA-loaded NMF appeared to be highly stable and well tolerated on human corneal epithelial cells (HCEC) and human retinal pigment epithelial cells (D407 cells). Overall, these studies suggest that TA in NMF is safe and suitable for human topical ocular drop application.

Transfer learning based classification of optical coherence tomography images with diabetic macular edema and dry age-related macular degeneration

We present an algorithm for identifying retinal pathologies given retinal optical coherence tomography (OCT) images. Our approach fine-tunes a pre-trained convolutional neural network (CNN), GoogLeNet, to improve its prediction capability (compared to random initialization training) and identifies salient responses during prediction to understand learned filter characteristics. We considered a data set containing subjects with diabetic macular edema, or dry age-related macular degeneration, or no pathology. The fine-tuned CNN could effectively identify pathologies in comparison to classical learning. Our algorithm aims to demonstrate that models trained on non-medical images can be fine-tuned for classifying OCT images with limited training data.

Restoration of Photoreceptors in Eyes with Diabetic Macular Edema

Purpose

To show the prognostic value of foveal microstructures using optical coherence tomography (OCT) for the restoration of inner segment/outer segment (IS/OS) junction layer following resolution of diabetic macular edema (DME).

Methods

Forty-one eyes of 39 patients with IS/OS damage at the time of DME that showed complete resolution of DME were included. Eyes were divided into 2 groups based on the IS/OS integrity at final visit, when edema was completely resolved: always damage group (damage at baseline and at final visit) and initial damage group (damage only at baseline). The OCT characteristics including the extent of the IS/OS damage, central subfield thickness (CST), maximum retinal thickness (MRT), presence or absence of subretinal fluid, duration of diabetic retinopathy, and duration of DME were studied. The integrity of IS/OS was evaluated at baseline and at last follow-up as percentage (0%-100%).

Results

Forty-four percent of eyes (18 eyes) achieved complete restoration of IS/OS after resolution of DME. There was no significant difference in CST or MRT during DME between the 2 groups. The always damage group had more IS/OS damage at baseline visit (23.6% ± 6.4% vs 10.7% ± 3.4%, p = 0.043) with a longer duration of DME (p = 0.025). Despite a borderline significance in visual acuity between the 2 groups at baseline (p = 0.05), the always damage group ended up with worse vision at last follow-up (p<0.001).

Conclusions

Patients with shorter duration of DME and less baseline IS/OS damage were more likely to have intact (restored) IS/OS after resolution of DME.

A Review on Recent Developments for Detection of Diabetic Retinopathy

Diabetic retinopathy is caused by the retinal micro vasculature which may be formed as a result of diabetes mellitus. Blindness may appear as a result of unchecked and severe cases of diabetic retinopathy. Manual inspection of fundus images to check morphological changes in microaneurysms, exudates, blood vessels, hemorrhages, and macula is a very time-consuming and tedious work. It can be made easily with the help of computer-aided system and intervariability for the observer. In this paper, several techniques for detecting microaneurysms, hemorrhages, and exudates are discussed for ultimate detection of nonproliferative diabetic retinopathy. Blood vessels detection techniques are also discussed for the diagnosis of proliferative diabetic retinopathy. Furthermore, the paper elaborates a discussion on the experiments accessed by authors for the detection of diabetic retinopathy. This work will be helpful for the researchers and technical persons who want to utilize the ongoing research in this area.

A Simple Spontaneous Vitreal Reflux Collecting Procedure During Intravitreal Injection: Set-Up and Validation Studies

AIM

To set-up a simple technique for collecting spontaneous vitreal reflux (VR) in patients undergoing intravitreal injection. Both total protein concentration and vascular endothelial growth factor (VEGF)/Interleukin 13 (IL13) levels were used to validate the technique.

METHODS

Sixty consecutive patients with neovascular age-related macular degeneration (nAMD, vitreal reflux drop, VR) and 10 patients underwent vitrectomy for macular hole (whole vitreous removal) were enrolled for the study as controls. Thirty-three out of 60 patients were also subjected to tear sampling. VR sampling was performed after the intravitreal injection. Four sampling tools (10 Schirmer strips, 10 microsponges, 20 millipore filters; 20 micropipettes) were tested. Analysis of protein concentration/composition was performed between VR samples and vitreous samples to analyze the difference. The concentration of VEGF and IL 13 levels between cases and control samples were compared.

RESULTS

Millipore and micropipette techniques allowed the collection of higher protein concentrations in VR samples, comparison of both protein concentrations revealed no significant difference in the protein profile. However, the micropipette sampling was found easier to perform and did not require additional protein extraction from a solid support (membrane). Indeed, tear proteins and drug contaminants were not detected in micropipette samples. Increased VEGF levels were detected in naive VR group and to a less extend in VR group of nAMD patients undergoing intravitreal injection, with respect to the controls (macular holes). No significant differences in IL13 levels were quantified in nAMD sub-groups, as compared to naive and controls.

CONCLUSIONS

Overall, we provide evidence for a safe method for sampling VR at the end of intravitreal injection. This procedure might represent an interesting approach either for the prognosis of disease or monitoring the efficacy of intravitreal therapy.

Efficacy of Intravitreal Triamcinolone Acetonide for Diabetic Macular Edema After Vitrectomy

PURPOSE

To compare the efficacy of intravitreal injection of triamcinolone acetonide (IVTA) for diabetic macular edema (DME) in vitrectomized eyes with DME without vitrectomy eyes.

METHODS

This retrospective comparative study evaluated the efficacy of IVTA treatment of DME in 26 consecutive eyes (23 patients). Changes in mean best-corrected visual acuity (VA) and mean central retinal thickness (CRT) were retrospectively evaluated before IVTA and during the 6-month period after IVTA.

RESULTS

Subjects were divided into 2 groups: 13 consecutive eyes (11 patients) with proliferative diabetic retinopathy or DME that underwent vitrectomy (vitrectomized group), and 13 consecutive eyes (12 patients) with DME who received IVTA, but did not undergo vitrectomy (nonvitrectomized group). In the vitrectomized group, there was a significantly decreased CRT for up to 4 months as compared to the thicknesses before IVTA. In the nonvitrectomized group, there was a significantly decreased CRT for up to 5 months after IVTA. In both groups, there was significant improvement in the VA for up to 4 months after IVTA.

CONCLUSION

IVTA may represent a valid treatment option for DME, even in vitrectomized eyes.

Ranibizumab 0.5 mg for Diabetic Macular Edema with Bimonthly Monitoring after a Phase of Initial Treatment: 18-Month, Multicenter, Phase IIIB RELIGHT Study

PURPOSE

To evaluate ranibizumab 0.5 mg using bimonthly monitoring and individualized re-treatment after monthly follow-up for 6 months in patients with visual impairment due to diabetic macular edema (DME).

DESIGN

A phase IIIb, 18-month, prospective, open-label, multicenter, single-arm study in the United Kingdom.

PARTICIPANTS

Participants (N = 109) with visual impairment due to DME.

METHODS

Participants received 3 initial monthly ranibizumab 0.5 mg injections (day 0 to month 2), followed by individualized best-corrected visual acuity (BCVA) and optical coherence tomography-guided re-treatment with monthly (months 3-5) and subsequent bimonthly follow-up (months 6-18). Laser was allowed after month 6.

MAIN OUTCOME MEASURES

Mean change in BCVA from baseline to month 12 (primary end point), mean change in BCVA and central retinal thickness (CRT) from baseline to month 18, gain of ≥10 and ≥15 letters, treatment exposure, and incidence of adverse events over 18 months.

RESULTS

Of 109 participants, 100 (91.7%) and 99 (90.8%) completed the 12 and 18 months of the study, respectively. The mean age was 63.7 years, the mean duration of DME was 40 months, and 77.1% of the participants had received prior laser treatment (study eye). At baseline, mean BCVA was 62.9 letters, 20% of patients had a baseline BCVA of >73 letters, and mean baseline CRT was 418.1 μm, with 32% of patients having a baseline CRT <300 μm. The mean change in BCVA from baseline to month 6 was +6.6 letters (95% confidence interval [CI], 4.9-8.3), and after institution of bimonthly treatment the mean change in BCVA at month 12 was +4.8 letters (95% CI, 2.9-6.7; P < 0.001) and +6.5 letters (95% CI, 4.2-8.8) at month 18. The proportion of participants gaining ≥10 and ≥15 letters was 24.8% and 13.8% at month 12 and 34.9% and 19.3% at month 18, respectively. Participants received a mean of 6.8 and 8.5 injections over 12 and 18 months, respectively. No new ocular or nonocular safety findings were observed during the study.

CONCLUSIONS

The BCVA gain achieved in the initial 6-month treatment period was maintained with an additional 12 months of bimonthly ranibizumab PRN treatment.

Ranibizumab for the treatment of degenerative ocular conditions

Degenerative ocular conditions, such as age-related macular degeneration, diabetic retinopathy, retinal vein occlusions, and myopic degeneration, have become a major public health problem and a leading cause of blindness in developed countries. Anti-vascular endothelial growth factor (VEGF) drugs seem to be an effective and safe treatment for these conditions. Ranibizumab, a humanized monoclonal antibody antigen-binding fragment, which inhibits all biologically active isoforms of VEGF-A, is still the gold standard treatment for the majority of these pathological entities. In this review, we present the results of the most important clinical trials concerning the efficacy and safety of ranibizumab for the treatment of degenerative ocular conditions.

Improvement of visual acuity based on optical coherence tomography patterns following intravitreal bevacizumab treatment in patients with diabetic macular edema

AIM

To report the visual outcome based on various patterns of optical coherence tomography (OCT) morphology in diabetic macular edema (DME), following treatment with anti-VEGF intravitreal bevacizumab (IVB) injection.

METHODS

Sixty-seven consecutive subjects with centre involving DME underwent intravitreal injection of Bevacizumab (1.25 mg/0.05 mL) in this retrospective, comparative, non randomized study. The DME was classified into one of four categories: focal, diffuse, focal cystoid and neurosensory detachment based on OCT. Best corrected visual acuity (BCVA), macular appearance, and OCT findings were used to decide whether the subject should have a repeat injection of intravitreal bevacizumab. Outcome measures were a change in mean BCVA (Snellen converted to logMAR) and central macular thickness (CMT) in each group during the six month follow-up period.

RESULTS

The mean BCVA improved to logMAR 0.23 at final follow-up from a baseline of 0.32 logMAR (P=0.040) in the focal group, logMAR 0.80 at final follow-up from a baseline of 0.82 logMAR (P=0.838) in the diffuse group, worsened to logMAR 0.53 at final follow-up from a baseline of 0.43 logMAR (P=0.276) in the focal cystoid group, and improved to logMAR 0.79 at final follow-up from a baseline of 0.93 logMAR (P=0.490) in the neurosensory detachment group. The mean CMT before treatment were 298.8±25.03 µm in the focal group, 310.8±40.6 µm in the diffuse group, 397.15±31.05 µm in the focal cystoid group and 401.03±75.1 µm in the neurosensory detachment group. A mean of 2.05 (range: 1-5) injections in the focal group, 1.32 (range: 1-2) in the diffuse group, 2.6 (range: 1-6) in the focal cystoid group and 2.6 (range: 1-6) in the neurosensory detachment group were performed during the six month follow-up period. Following intravitreal bevacizumab treatment, vision improved, remained unchanged or worsened in 11, 7 and 2 subjects in focal group; 11, 9 and 8 in diffuse group; 0, 2 and 4 in focal cystoid group and 5, 5 and 3 subjects respectively in neurosensory detachment group.

CONCLUSION

OCT morpholgy patterns in DME may predict the effects of intravitreal bevacizumab treatment, and patients with focal DME are most likely to benefit from the improvent of visual acuity from this treatment.