A Systematic Review and Meta-Analysis of Clinical Outcomes of Intravitreal Anti-VEGF Agent Treatment Immediately after Cataract Surgery for Patients with Diabetic Retinopathy

Aims To examine possible benefits of intravitreal anti-vascular endothelial growth factor (VEGF) agent treatment immediately after cataract surgery for patients with diabetic retinopathy (DR). Methods A comprehensive literature search was performed using the Cochrane collaboration methodology to identify randomized controlled trials (RCTs) and comparative studies of cataract surgery with or without anti-VEGF agent treatment for any diabetic retinopathy. Meta-analyses were performed for clinical outcome parameters including changes in macular thickness (MT), best-corrected visual acuity (BCVA), incidence of diabetic retinopathy and maculopathy progression, laser treatment rate, and other complications. Results Nine RCTs and 3 nonrandomized comparative studies were identified and used for comparing cataract surgery with intravitreal bevacizumab (IVB) or intravitreal ranibizumab (IVR) treatment (338 eyes, intervention group) to cataract surgery alone (329 eyes, control group). Analysis of all data showed that the mean BCVA at 1 week postoperatively had no statistically significant difference in the two groups, but at 1, 3, and 6 months postoperatively, the mean BCVA was statistically significantly better in the anti-VEGF treatment group than that in cataract surgery alone group. Analysis of all data showed that the mean MT was statistically significantly less in the anti-VEGF treatment group at 1 week and 1, 3, and 6 months postoperatively (P=0.05, P=0.006, P=0.0001, and P=0.0001, respectively); but postoperative clinical outcomes were differentiated from the type of anti-VEGF agents, IVB or IVR, and the existing macular edema preoperatively. Intravitreal anti-VEGF agent treatment statistically significantly reduced the incidence of diabetic retinopathy progression and maculopathy progression compared to the control group (P=0.0003, P < 0.00001, respectively). Conclusion IVB or IVR treatment immediately after cataract surgery may represent a safe and effective strategy to prevent postoperative macular thickening or reduce macular edema and result in greater mean improvements in visual acuity for diabetic patients.


Introduction
Diabetic macular edema (DME) is a major cause of vision loss for diabetic patients [1,2]. Diabetic patients have a higher prevalence of early cataract need for surgery. e evidences revealed that intraocular level of inflammatory cytokines and VEGF in eyes with diabetic retinopathy can be further increased by cataract surgery and significantly increase the risk of DME, which was different from cystoid macular edema (CME) following cataract surgery (Irvine-Gass syndrome) for nondiabetic patients [3][4][5]. Topical nonsteroidal anti-inflammatory drugs (NSAIDs), either as monotherapy or combined with topical corticosteroids, appear to be effective for the prophylaxis and treatment of CME, but the intravitreal administration of anti-VEGF or steroids, at the moment of cataract surgery, seems to be a feasible solution to prevent or cure diabetic macular edema [6,7].
In an attempt to detect benefits in safety and efficacy as the primary comparative criteria, we performed a systematic review and meta-analysis of existing randomized controlled trials (RCTs) and high-quality comparative studies of cataract surgery with or without intravitreal anti-VEGF agent treatment for the treatment of DR including nonproliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), or clinically significant macular edema (CSME).

Materials and Methods
is meta-analysis was performed according to a predetermined protocol described previously [23,24].

Literature Search.
Two reviewers independently searched the following electronic databases: PubMed, EMBASE, and Wanfang Data (e-resources for China studies) and the Cochrane Controlled Trials Register up to June 30, 2018. For maximum sensitivity, we used free text and thesaurus terms including "cataract surgery or phacoemulsification," "diabetic retinopathy or diabetic macular edema," and "bevacizumab or Avastin or ranibizumab or Lucentis or aflibercept or anti-vascular endothelial growth factor agents." All published RCTs and comparative studies comparing cataract surgery alone versus cataract surgery with intravitreal anti-VEGF agent treatment for diabetic retinopathy were included. Patients preparing for cataract surgery were presented with diabetic retinopathy and greater than 18 years. At least one or more clinical outcomes representing clinical outcome parameters must be assessed and published.
ere was no language restriction on the publications. e selected studies were appraised by two reviewers, who independently assessed their quality using the methods recommended in the Cochrane Handbook for Systematic Reviews of Interventions [25].
Discordance about study inclusion and quality assessment between the two reviewers was resolved through discussion until 100% agreement was reached on the final interpretation of the data.

Outcome Measure.
e clinical outcome parameters included mean central macular thickness (CMT), bestcorrected visual acuity (BCVA), incidence of diabetic retinopathy and maculopathy progression, incidence of neovascular glaucoma (NVG), and rate of laser photocoagulation treatment postoperatively. Other adverse events such as elevation of intraocular pressure (IOP), retinal detachment, and ocular inflammation were all recorded.

Data Extraction and Analysis.
e studies were tabulated and methodologically evaluated to assess homogeneity. Any heterogeneity between the studies would not be justified to pool the assessed outcomes. A customized data extraction form, as described in the Cochrane Handbook for Systematic Reviews of Interventions, was used to record the duration of the trial, sample size, dropouts, the system and ocular baseline features, the inclusion and exclusion criterion for patients, the dosage and location of intravitreal anti-VEGF agent injection, and postoperative treatment.

Statistical Analysis.
Quantitative data were analyzed using the Cochrane Review Manager (RevMan) version 5.0 software. Summary estimates, including 95% confidence interval (CI), were calculated. For continuous outcome data (e.g., mean macular thickness) means and standard deviations were used to calculate a weighted mean difference (WMD). For dichotomous outcomes (e.g., incidence of retinopathy progression), the odds ratio (OR) was calculated.
Statistical heterogeneity was tested using Q statistic of chi-square value test and I 2 test. Fixed effects models were used, unless significant evidence of statistical heterogeneity or clinical diversity was found. For results showing significant heterogeneity (I 2 > 50%), random-effects meta-analysis was performed. Outcome measures were assessed on an intent-to-treat (ITT) basis. A P value less than 0.05 was considered statistically significant. Subgroup analysis was performed according to the inclusion criterion.
Five studies included NPDR patients with no preexisting DME and evaluated intravitreal anti-VEGF agents at cataract surgery for prevention of postoperative DME [12,16,17,19,22]; five studies included NPDR patients with preexisting DME and evaluated intravitreal anti-VEGF agents at cataract surgery for management of diabetic maculopathy [13,14,18,20,21]. Two studies included NPDR or PDR patients with or without CSME and evaluated intravitreal anti-VEGF agents at cataract surgery for management of diabetic maculopathy [11,15].
Additive other treatments, such as intravitreal anti-VEGF agent treatment, were not performed during the follow-up period in the all included studies. e present meta-analysis involved 338 eyes receiving cataract surgery with anti-VEGF agent treatment and 329 eyes treated with cataract surgery alone. e selection of 12 studies is summarized in Tables 1 and 2.

Macular ickness.
For NPDR without DME, one study [12] applying IVB revealed that the mean MTat 1 month and 6 months postoperatively was statistically significantly less in the IVB treatment group than the control group; 4 studies [16,17,19,22] applying IVR revealed that the MT at 1 week and 1, 3, and 6 months postoperatively was statistically significantly less in the IVR treatment group (Table 3). All data showed that change of postoperative MTfrom 1 week to 6 months compared with the baseline was minor or increased little in the anti-VEGF treatment group (from −1 to +24 μm), but there was apparent increase of postoperative MT compared with the baseline in the control group (from +12 to +90 μm).
For NPDR with DME, 3 studies [13,14,21] applying IVB revealed that the mean MT at 1, 3, and 6 months postoperatively was statistically significantly less in the IVB treatment group than the control group; 2 studies [18,20] applying IVR revealed that the MT at 1 week and 1, 3, and 6 months postoperatively was statistically significantly less in the IVR treatment group (Table 3). All data showed that change of postoperative MT from 1 week to 6 months compared with the baseline was minor or decreased apparently in the anti-VEGF treatment group (from +10 to −98 μm), but there was apparent increase of postoperative MT compared with the baseline in the control group (from −22 to +77 μm).
Analysis of all data showed that the mean MT was statistically significantly less in the anti-VEGF treatment group at 1 week and 1, 3, and 6 months postoperatively (Table 3).

BCVA.
For NPDR without DME, one study [12] applying IVB revealed that the mean BCVA was statistically significantly better in the IVB treatment group than that in the cataract surgery alone group at 1 month postoperatively, but the mean BCVA at 3 and 6 months postoperatively had no statistically significant difference in the two groups. ree studies [16,17,19] applying IVR revealed that the mean BCVA at 1 week postoperatively had no statistically significant difference in the two groups. At 1 and 3 months postoperatively, 1 study [16] reporting LogMAR visual acuity revealed that the mean BCVA had no statistically significant difference in the two groups, and 1 study [19] reporting Snellen visual acuity revealed that the mean BCVA was statistically significantly better in the IVB treatment group. At 6 months postoperatively, the mean BCVA of two studies [16,17] was statistically significantly better in the IVR treatment group (Table 4).
For NPDR with DME, 3 studies [13,14,21] applying IVB revealed that the mean BCVA at 1 week postoperatively had no statistically significant difference in the two groups, but at 1, 3, and 6 months postoperatively, the mean BCVA was statistically significantly better in the IVB treatment group than that in the cataract surgery alone group. Two studies [18,20] applying IVR revealed that the mean BCVA at 1 week and 1, 3, and 6 months postoperatively was statistically significantly better in the IVR treatment group (Table 4).
Analysis of all data showed that the mean BCVA at 1 week postoperatively had no statistically significant difference in the two groups, but at 1, 3, and 6 months postoperatively, the mean BCVA was statistically significantly better in the anti-VEGF treatment group than that in the cataract surgery alone group (Table 4).

Postoperative Complications and Treatment.
e incidence of retinopathy and maculopathy progression after cataract surgery were statistically significantly less in the intravitreal anti-VEGF treatment group (P � 0.0003, P < 0.00001, respectively) than those in the control group (Table 5). e incidence of NVG progression had no statistically significant difference between both groups, but had a trend to statistical significance (P � 0.07) ( Table 5). e rate of laser photocoagulation treatment had no statistically significant difference between both groups (P � 0.72) ( Table 5).
ere was no significant increase of intraocular pressure after surgery in each group (P � 0.21) (Table 5). e incidence of adverse events that were related to the injection itself, such as vitreous hemorrhage and conjunctival hemorrhage, had no statistically significant      Journal of Ophthalmology difference between both groups (P � 0.97, P � 0.28, respectively) ( Table 5).
No adverse events, such as retinal detachment, severe ocular inflammation, endophthalmitis, or systemic adverse, were reported during the follow-up periods.

Discussion
e meta-analysis results showed that the intravitreal anti-VEGF agent treatment, irrespective of IVB or IVR, immediately after cataract surgery could prevent the increase in macular thickness of eyes with NPDR without existing DME and decreasing macular thickness in NPDR with existing DME. is effect seemed to hold in 6 months postoperatively.
Similarly, the meta-analysis data also showed the intravitreal anti-VEGF agent treatment could improve better postoperative visual acuity of eyes with NPDR. For NPDR without existing DME, the effect of visual acuity improvement by IVB seemed to hold in one month postoperatively, and the effect of IVR could hold in one to 6 months postoperatively. For NPDR with existing DME, the effect of visual acuity improvement by IVB seemed to hold in 1 to 6 months postoperatively, and the effect of IVR could hold in one week to 6 months postoperatively.
Although there was no study to directly compare IVB to IVR for the treatment of diabetic macular edema immediately after cataract surgery. In the future, more trials will be needed to verify whether IVR was superior to IVB and the therapy effect of IVR was more quickly and more durable for DME after cataract surgery. Many clinical trials compared visual acuity and OCT outcomes associated with IVB vs IVR for the management of DME [26,27]. IVB and IVR were associated with similar effects on central macular thickness in patients with DME. IVR is associated with greater improvement in BCVA at some study visits, and the mean number of injections is higher in the IVB group.
For subgroups of eyes mixed CSME, NPDR, or PDR with no difference in postoperative macular thickness from 1 month to 3 months, the consequential effect of laser photocoagulation ought to be considered [11,15]. e reported rate of laser photocoagulation treatment after cataract surgery had no significant difference between the both groups [11,15]. is reflected that the ocular baseline especially the fundus was similar between both groups because the included studies reported that laser photocoagulation for CSME and PDR was performed promptly after cataract surgery in both groups based on ETDRS criterion. Laser photocoagulation is still a standard therapy for DME and has proven efficacy in large clinical trials [28,29]. So, the early macular edema or obvious macular edema was treated. Both groups had similar central macular thickness and visual acuity during the 3-month follow-up period. May be the IVB beneficial effect for macular edema was not quick. However, the 6-month data revealed that the IVB treatment still had a little beneficial effect in preventing the macular thickening.
Although the treatment rate of laser photocoagulation had no significant difference, we thought the efficacy of laser photocoagulation after the intravitreous pretreatment group was high. As Lang's study revealed, intravitreal anti-VEGF agent treatment plus laser has also proven to be more Journal of Ophthalmology 7 effective for the treatment of PDR compared to laser alone [30]. It can improve the resolution of vitreous and retinal hemorrhage and facilitate laser photocoagulation completion. So, the completely laser therapy significantly would reduce the incidence of progression of retinopathy and maculopathy in eyes that received intravitreal anti-VEGF agent treatment. is is a synergistic prophylactic effect determined by laser therapy and intravitreal anti-VEGF agent treatment. Although the incidence of progression of NVG had no significant difference in both groups, the incidence (7/63) of progression of NVG in eyes without receiving intravitreal anti-VEGF agent treatment was more times higher than those (1/62) with receiving treatment, and there was a trend to statistically significant difference.  Many studies have been published to prove whether cataract surgery itself or other factors increase the risk of DR progression postoperatively [31,32]. Whatever reasons, cataract surgery must be performed for some diabetic patients. And how to prevent postoperatively macular thickening and retinopathy progression is the main question. Besides the beneficial effect of intravitreal anti-VEGF agent treatment, the traditional treatment still yielded better results in terms of macular thickness reduction and improvement of VA, such as laser photocoagulation, intravitreal triamcinolone, or dexamethasone implant treatment [7,[33][34][35], but intravitreal triamcinolone treatment was associated with a greater risk of intraocular pressure elevation. An optional therapy was determined by not only the better clinical outcomes but also the fewer side effects. e selected trials revealed that intravitreal anti-VEGF agent treatment was well tolerated, and no local and systemic adverse events were noticed during the study. However, from the values of macular thickness during follow-up, there was a trend of macular thickening from 3 to 6 months postoperatively [15,16]. e drawback of its nonpermanent effect and higher costs ought to be of concern [36]. Further therapy for diabetic retinopathy after cataract surgery was integrated by different treatment methods according to their respective advantages.
is present meta-analysis might have some limitations. One weakness of our study was that there was no trial exclusively focused on PDR as subgroup analysis. e metaanalysis results should be accurate and comprehensive if adding the subgroup analysis of PDR with or without macular edema. Secondly, the studies included were heterogeneous in terms of study location, population, number of patients from different studies and basal condition, and study quality (relative methodological strengths and weaknesses). e strengths of this study include the randomized study design and nonrandomized comparative study design. e baseline conditions in these nonrandomized comparative studies were highly controlled. So it is feasible to include these studies to perform meta-analysis. More study participants were included to research, and the results were more closed to be representative of the real-world population with DME. ese findings still need to be interpreted with caution. Bevacizumab is an off-label anti-VEGF agent for DME or postsurgical macular edema. It must be used with caution.
In conclusion, the present study demonstrated that intravitreal anti-VEGF agent treatment plus cataract surgery may represent a safe and effective strategy to prevent postoperative macular thickening or decreasing macular edema and result in greater mean improvements in visual acuity for diabetic patients. In future, new treatments, such as other anti-VEGF agent injection and intravitreal anti-VEGF agents combined subtenon triamcinolone injection, should be investigated to improve clinical efficacy, and high-quality, largescale, multicenter randomized control trials will be needed to verify [37]. e optimized therapy should be suggested when using the data from the evidence-based study to guide treatment considerations for an individual patient.

Conflicts of Interest
e authors declare that they have no conflicts of interest.