Safety of Receiving Anti–Vascular Endothelial Growth Factor Intravitreal Injection in Office-Based vs Operating Room Settings: A Meta-analysis

Tong Li; Junran Sun; Jingyu Min; Shuangwen Zhou; Xiaolin Zhu; Huixun Jia; Xiaodong Sun

doi:10.1001/jamaophthalmol.2021.3096

. 2021 Aug 19;139(10):1–10. doi: 10.1001/jamaophthalmol.2021.3096

Safety of Receiving Anti–Vascular Endothelial Growth Factor Intravitreal Injection in Office-Based vs Operating Room Settings

A Meta-analysis

Tong Li ^1,², Junran Sun ^1,², Jingyu Min ³, Shuangwen Zhou ³, Xiaolin Zhu ³, Huixun Jia ^1,^2,^4,^✉, Xiaodong Sun ^1,^2,^4,⁵

¹Department of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

²National Clinical Research Center for Ophthalmic Diseases, Shanghai, China

³Novartis Pharmaceuticals, Shanghai, China

⁴Shanghai Engineering Center for Visual Science and Photomedicine, Shanghai, China

⁵Shanghai Key Laboratory of Fundus Diseases, Shanghai, China

Accepted for Publication: June 17, 2021.

Published Online: August 19, 2021. doi:10.1001/jamaophthalmol.2021.3096

^✉

Corresponding Author: Huixun Jia, BS, Department of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, 100 Haining Rd, Shanghai 200080, China (jiahuixun@163.com).

Author Contributions: Ms Jia and Dr X. Sun had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Li and J. Sun contributed equally.

Study concept and design: Zhu, Jia, X. Sun.

Acquisition, analysis, or interpretation of data: Li, J. Sun, Min, Zhou, Jia, X. Sun.

Drafting of the manuscript: Li, J. Sun, Min, Zhou.

Critical revision of the manuscript for important intellectual content: Zhu, Jia, X. Sun.

Statistical analysis: Li, Jia.

Administrative, technical, or material support: Min, Zhou, Zhu.

Study supervision: Zhu, Jia, X. Sun.

Conflict of Interest Disclosures: Dr Li has received grants from the Science and Technology Commission of Shanghai Municipality. Dr J. Sun has received grants from the National Nature Science Foundation of China. Ms Jia has received grants from the Scientific Project of Shanghai Municipal Health Commission and the National Nature Science Foundation of China. No other disclosures were reported.

Funding/Support: This work was supported by grant 2019ZX09301113 from the National Major Scientific and Technological Special Project for “Significant New Drugs Development” during the Thirtieth Five-year Plan Period (Dr X. Sun), grant 81730026 from the National Nature Science Foundation of China (Dr X. Sun), grants 19495800700 (Dr X. Sun) and 19YF1439800 (Dr Li) from the Science and Technology Commission of Shanghai Municipality, and grant 201940151 from the Scientific Project of Shanghai Municipal Health Commission (Ms Jia).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We thank Yuan Li, MSc (Novartis Pharmaceuticals, Shanghai, China), for providing medical writing support on the manuscript, in accordance with Good Publication Practice guidelines. No financial compensation was provided.

^✉

Corresponding author.

PMCID: PMC8377608 PMID: 34410325

Key Points

Question

Are the safety outcomes of intravitreal injection with anti–vascular endothelial growth factor agents in the office-based setting comparable with outcomes in the operating room?

Findings

In this meta-analysis including 31 studies with a total of 1 275 815 injections, no difference between rates of postinjection endophthalmitis were identified in the 2 settings. The rates of other ocular adverse events were quite low, with no systemic adverse events reported.

Meaning

This study cannot determine whether performing injection in the operating room is more appropriate in lower-income regions.

This meta-analysis evaluates the safety outcomes of intravitreal injection with anti–vascular endothelial growth factor agents in the operating room vs office-based setting.

Abstract

Importance

Compared with the operating room (OR), office-based intravitreal injection (IVI) is considered a more cost-effective and convenient approach, yet clinical outcomes of IVIs with anti–vascular endothelial growth factor (VEGF) agents in different settings (office-based vs OR) have not been systematically evaluated.

Objective

To evaluate the safety outcomes of IVI with anti-VEGF agents in the OR vs office-based setting.

Data Sources

PubMed, Embase, Cochrane Library, Web of Science, and ClinicalTrials.gov were searched from inception to July 2020.

Study Selection

Eligible studies reporting on patients who received IVIs with anti-VEGF drugs with a clearly stated injection setting of the office or OR.

Data Extraction and Synthesis

Two reviewers independently screened studies, extracted data, and assessed risk of bias. A meta-analysis was conducted to determine the rates of endophthalmitis (EO) and culture-positive EO.

Main Outcomes and Measures

Rates of EO and culture-positive EO following anti-VEGF IVIs in the OR and office-based setting.

Results

Thirty-one studies with a total of 1 275 815 injections were included. Comparative analysis suggested no difference between rates of EO after IVIs performed in the office and OR settings (odds ratio, 3.06; 95% CI, 0.07-139.75; P = .57; I² = 80%) were identified, yet a higher rate of culture-positive EO was found in the office setting (odds ratio, 21.52; 95% CI, 2.39-193.55; P = .006; I² = 0%). The pooled rates of EO following anti-VEGF IVIs were 0.03% (95% CI, 0.03-0.04) and 0.02% (95% CI, 0.01-0.04) in office and OR settings, respectively, and the pooled rates of culture-positive EO were 0.01% (95% CI, 0.01-0.02) and 0.01% (95% CI, 0-0.02). The pooled rates of other ocular and systemic adverse events were low.

Conclusions and Relevance

The rate of clinically suspected or culture-positive EO following anti-VEGF IVIs was low whether the procedure was performed in the office or OR setting. Bacterial spectrum could differ between the 2 settings. This meta-analysis could not determine if it is more appropriate to give treatment in the OR for safety reasons in low-income compared with higher-income regions in the world.

Introduction

First described at the turn of the 20th century for repairing retinal detachments (RDs), intravitreal injections (IVIs) have become the most widely performed procedures in the ophthalmic field.¹ IVIs with anti–vascular endothelial growth factor (VEGF) agents have significantly improved management and visual prognosis of angiogenic retinal diseases, including neovascular age-related macular degeneration, diabetic macular edema, retinal vein occlusion, proliferative diabetic retinopathy, choroidal neovascularization, and other visual impairments.²

Endophthalmitis (EO) is a severe form of eye inflammation that may result in irreversible blindness if the intraocular infection is not properly treated.³ To our knowledge, the largest meta-analysis on EO to date revealed an overall rate of 0.06% (197 of 350 535 injections) following IVIs with all anti-VEGF agents.⁴ Anti-VEGF injections require repeated injections on a regular basis.⁵ Despite that subsequent injections did not increase the risk of EO compared with the first one, awareness of the risk remains clinically relevant as the number of injections increases.⁴

IVIs with anti-VEGF agents are predominantly performed in the operating room (OR) in Europe and in resource-limited countries, such as China⁶ and India,⁷ owing to the perception of a decreased risk of infection.⁸ Given the high volume of IVIs performed annually with limited availability of ORs, these injections were typically administered in an office setting in the US.^9,10,11 Iatrogenic infection remains a primary concern for office-based IVI,¹² yet evidence from primary research remains controversial; while Abell et al¹³ found a significantly higher risk of EO following office-based IVIs than in the OR (relative risk, 13.1; 95% CI, 1.5-112.4), no significant difference was observed between the 2 settings in the study by Tabandeh et al⁸ (odds ratio, 0.53; 95% CI, 0.09-3.18). These findings suggest a need for systemic evidence on EO rates and other safety outcomes following anti-VEGF IVIs in the 2 settings. Here, we conducted a meta-analysis evaluating the safety of anti-VEGF IVI in the office and OR settings and discuss the implications for future practice.

Methods

This study was performed based on the Meta-analysis of Observational Studies in Epidemiology (MOOSE) reporting guideline.¹⁴

Literature Search Strategies

PubMed, Embase, Cochrane Library, Web of Science, and ClinicalTrials.gov were searched from database inception to July 2020. The literature search was restricted to English-language articles using a combination of subject wording, keywords, and free-text terms, including Lucentis, ranibizumab, Eylea, aflibercept, conbercept, Beovu, brolucizumab, bevacizumab, anti-VEGF, operating room, office-based, intravitreal injection, and IVI. Details of the search are shown in the eMethods in the Supplement. Bibliographic searches of relevant reviews, guidelines, and meta-analysis were conducted to identify additional studies.

Eligibility Criteria

A study was eligible for inclusion if the following criteria were met: (1) the study focused on patients who received IVI with anti-VEGF drugs alone or combined with panretinal photocoagulation, laser, or photodynamic therapy; (2) there was a clearly stated injection setting of office or OR; (3) the article reported the safety outcomes following anti-VEGF IVIs; and (4) the study was primary clinical research, including randomized clinical trials (RCTs), nonrandomized comparative cohorts, cross-sectional studies, and case series.

Study Selection

Two reviewers (T.L. and J.S.) independently screened the titles and abstracts to assess the initial eligibility of a study. Full texts of initially eligible studies were retrieved for detailed inspection. Discrepancies between the 2 reviewers were resolved through discussion and consultation with a senior reviewer (H.J.) when necessary. Studies that were not omitted after full-text screening were considered suitable for inclusion.

Data Extraction and Quality Assessment

Variables extracted included (1) study characteristics (first author, publication year, study period, geographic location, number of patients/eyes, age, sex, dominant indication, study design, and follow-up duration), (2) intervention and/or comparison (anti-VEGF drug, injection settings, injection number, same-day bilateral injection, sterile condition, and drug preparation and doses), and (3) details of a specific safety outcome. The methodological quality was evaluated according to the design of each study included: RCTs were evaluated using the revised Cochrane risk-of-bias tool (https://www.riskofbias.info), case series were assessed with the National Institute for Clinical Excellence criteria,¹⁵ and case-control and comparative cohort studies were assessed with the Newcastle-Ottawa Scale.¹⁶

Outcome Measures

Safety outcomes of this study included rates of EO, culture-positive EO, RD, vitreous hemorrhage, intraocular pressure elevation, intraocular inflammation, and other ocular and systemic adverse events (AEs). Each outcome was summarized according to the settings.

Statistical Analysis

Meta-analysis was performed when 2 or more studies reported the same outcome; otherwise, results were separately described for any study with a different definition of outcomes. Quantitative data analysis was conducted using R software version 4.0.2, META package (The R Foundation). Peto odds ratio method was used for analysis of dichotomous variables when event rates were less than 1%; otherwise, Mantel-Haenszel method was used.¹⁷ Considering that the risk of EO might be associated with number of injections, rates of EO were combined with number of EO out of the total number of injections, which is in line with previous studies.^10,18 Generalized linear mixed models with the logit link were used for pooled analysis of single proportions to account for within-study uncertainties.^19,20

Heterogeneity across the studies was examined using I² statistics. The estimates were pooled with a fixed-effect model when no significant heterogeneity (I² less than 50%) was observed; otherwise, a random-effect model was used. Sensitivity analysis was performed evaluating the impact of country income level on injection safety.²¹ Publication bias was assessed by Egger test for any outcome with 10 or more studies included. Q test was used to calculate P values for heterogeneity, while z test was used for overall effect. Two-sided P values less than .05 were considered statistically significant.

Results

Study Selection and Characteristics

Figure 1 shows the selection process of studies identified through the database search, with 262 initial records and 9 additional records through other sources. Ultimately, 31 studies with a total of 1 275 815 injections were included for qualitative synthesis.

The baseline characteristics of the included studies are detailed in eTable 1 in the Supplement. Seventeen studies reported a total of 944 765 injections administered in the office setting,^{8,13,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36} while 14 studies reported 302 039 injections administered in the OR.^{6,7,8,13,37,38,39,40,41,42,43,44,45,46} Only 2 studies directly compared office-based injections with the OR-based injections: Abell et al¹³ reported no significant difference in age, sex, ocular comorbidities, or socioeconomic status between the 2 injection settings, while Tabandeh et al⁸ indicated similar baseline diagnosis; patients in both studies were injected with ranibizumab or bevacizumab. A total of 28 studies (90%) reported low risk of bias in 50% or more items in methodological quality assessment (eTable 2 in the Supplement). Prophylaxis measures for postinjection infection are shown in eTable 3 in the Supplement.

Overall Rates of EO

For comparative analysis (Figure 2A),^8,13 there was no difference between the 2 injection settings (odds ratio, 3.06; 95% CI, 0.07-139.75; P = .57), but significant heterogeneity was observed (I² = 80%). The pooled rates of EO in the office and OR settings are presented in Figure 3.^{6,7,8,13,22,23,25,26,27,28,29,30,31,32,33,34,35,36,38,39,40,41,43,45,46} In the office setting, the rates of EO following IVI of total anti-VEGF drugs, ranibizumab, bevacizumab, and aflibercept were 0.03% (95% CI, 0.03-0.04), 0.03% (95% CI, 0.03-0.04), 0.04% (95% CI, 0.03-0.05), and 0.04% (95% CI, 0.02-0.05), respectively. In the OR setting, the rates of EO following IVI of total anti-VEGF drugs, ranibizumab, and bevacizumab were 0.02% (95% CI, 0.01-0.04), 0.02% (95% CI, 0-0.06), and 0.04% (95% CI, 0.01-0.13), respectively. There was no significant heterogeneity for total anti-VEGF drugs, ranibizumab, bevacizumab, and aflibercept in the office setting (I² = 8%; I² = 29%; I² = 0%; I² = 0%, respectively), while statistical heterogeneity was found for all 3 groups in the OR setting (I² = 85%; I² = 81%; I² = 61%). Additionally, 2 studies reported no EO following 358 IVIs of aflibercept in the OR.^41,45

Rates of Culture-Positive EO

The culture-positive EO rate was significantly higher in the office setting compared with the OR setting, with no statistical heterogeneity being observed (odds ratio, 21.52; 95% CI, 2.39-193.55; P = .006; I² = 0%) (Figure 2B). The pooled rates of culture-positive EO in the office and OR settings are presented in Figure 4.^{6,7,8,13,25,27,29,30,31,32,34,35,36,38,39,40,41,43,46} The rates of culture-positive EO following IVI of total anti-VEGF drugs, ranibizumab, and bevacizumab were 0.01% (95% CI, 0.01-0.02), 0.01% (95% CI, 0.01-0.02), and 0.01% (95% CI, 0.01-0.02), respectively, in the office setting with no statistical heterogeneity (I² = 0%; I² = 0%; I² = 0%). In the OR setting, rates of culture-positive EO following IVI of total anti-VEGF drugs, ranibizumab, and bevacizumab were 0.01% (95% CI, 0-0.02), 0.01% (95% CI, 0-0.03), and 0.02% (95% CI, 0-0.12), respectively, with significant heterogeneity (I² = 79%; I² = 51%; I² = 64%).

Microbiologic Spectrum of Culture-Positive EO

Sixteen studies reported a total of 11 microbial species accounting for the EO (Figure 5).^{7,8,13,25,29,30,31,32,34,35,36,38,39,40,43,46} In the office setting, the 2 most common pathogens were Coagulase-negative staphylococci (CoNS) and Streptococcus viridans (46.5% [53 of 114] and 25.4% [29 of 114], respectively), whereas in the OR, the predominant pathogens were CoNS and Staphylococcus aureus (57% [13 of 23] and 26% [6 of 23], respectively).

Sensitivity Analysis

The only study in the office setting of a resource-limited country had the highest EO rate across all office-based studies.²⁸ Removing this study did not significantly alter the overall effect and heterogeneity. For studies in the OR, EO rates were overall higher in resource-limited countries than higher-income ones (eFigures 1 to 5 in the Supplement).

Other Ocular and Systemic AEs

No patient developed RD according to 2 studies^23,27 in the office setting and 1 study⁶ in the OR, whereas another study⁴⁰ reported an RD rate of 0.003% in the OR. Two studies in the office setting^23,27 gave a combined incidence rate of 0.01% and 0.02% for vitreous hemorrhage and RT, respectively. One case of cataract (0.003%) occurred in the OR,⁴⁰ but none occurred in the office setting.²⁷ Twenty cases of noninfectious inflammation were reported following 5356 aflibercept injections administered in 7 office locations in the US.²⁴ In total, 5 studies,^{6,23,27,42,44} including 2 in the office and 3 in the OR, claimed no systemic complications.

Publication Bias

No publication bias was indicated for outcomes with 10 or more studies included (eFigure 6 in the Supplement).

Discussion

IVIs of anti-VEGF agents are conducted in the OR because of the perception that the negative pressure environment of the OR can effectively reduce the risk of intraocular infections.^38,47 However, a growing number of practitioners in high-income countries, such as the US, UK, Japan, Australia, and Canada, began to perform office-based IVIs as it is a more cost-effective, convenient, and efficient approach without compromising safety.^{10,11,13,23,24,25,33,34,47,48,49} In these 25 studies, we found an overall EO rate of 0.03% (95% CI, 0.03-0.04) in the office setting and 0.02% (95% CI, 0.01-0.04) in the OR. McCannel¹⁸ reported an overall EO rate of 0.05% (95% CI, 0.04-0.07) following office-based anti-VEGF IVIs. Sigford et al¹¹ used EO rates in the US (0.05%) and Europe (0.03%) as surrogates to the estimated rates in the office and OR settings, respectively. Through pooling the studies with a well-described setting, our results were in general comparable with the previous reports.

Our analysis revealed a culture-positive EO rate of 0.01% (95% CI, 0.01-0.02) in the office setting and 0.01% (95% CI, 0-0.02) in the OR setting. Our comparative analysis, however, observed that culture-positive EO rates were significantly elevated in the office setting, yet this finding may be overestimated, as socioeconomic status was a potential confounder of the injection setting in the study by Abell et al.¹³ In line with previous studies,^18,50 CoNS accounted for most culture-positive EO cases in both settings. As a normal flora typically found on the skin and conjunctiva, the spread of CoNS during the injection procedure (that is, from a patient’s conjunctiva and lacrimal apparatus to the vitreous body) is a notable cause of infectious EO.⁵¹ Almost all studies included in our analysis reported povidone-iodine use prior to administering an IVI. Application of topical povidone-iodine for at least 30 seconds before each IVI can effectively reduce conjunctival bacteria counts and has become the standard of care for prophylaxis of intraocular infection.^9,47,52 S viridans was predominantly isolated in the office settings, which is consistent with findings from Busch et al² that a decreased rate of Streptococcus-induced EO was reported after IVIs in the OR. S viridans is a normal bacterial flora typically found in the oral cavity, upper respiratory tract, and gastrointestinal tract.⁵¹ It is postulated that talking or coughing during the IVI procedure, especially in the office setting where a face mask is not a standard of care, increases the chance of needle or conjunctiva contamination.^13,50,51 Most of our studies in the OR reported using sterile gloves, masks, speculum, drape, and hand antisepsis, but many studies in the office setting either did not report or reported nonadequate use of 1 or more of these aseptic techniques. As the vitreous is an immune-privileged site susceptible to infection even under a small inoculum of low virulence bacteria, adherence to masking and silence and other aseptic measures throughout the procedure are warranted to decrease bacterial dispersion.^18,53,54

Young age⁵⁵ and female sex^55,56,57 have been identified as risk factors associated with EO after IVIs of anti-VEGF medications. However, more than two-thirds of the included studies did not report patients’ age or sex (eTable 1 in the Supplement), thus restricting us from further analyzing their impact within each injection setting (OR vs office). In a retrospective cohort of 818 558 IVIs, Kiss et al⁴ found a more than 2-fold elevated risk of EO associated with aflibercept than ranibizumab, yet our pooled EO rates were not different between aflibercept, ranibizumab, and bevacizumab. Sterile inflammation has been reported in patients receiving aflibercept IVIs.^24,58 Since noninfectious and infectious EO may share similar symptoms, such as pain, conjunctival infection, and hypopyon, differential diagnosis can be difficult and at the discretion of the treating ophthalmologist.^29,58 Therefore, different diagnostic criteria may account for the various EO rates reported after aflibercept injection.

Overall and culture-positive EO rates for ranibizumab appear higher in the office setting, while rates for bevacizumab seem higher in the OR, yet none of these results were statistically meaningful. Our findings give the implication that not only a clean environment with strict aseptic rules during the injection procedure but also the degree to which sterile techniques were used for syringe preparation may also affect postinjection EO. Ranibizumab, initially available as single-dose vials, received its first prefilled syringe approval by the European Union in 2013 and the US Food and Drug Administration in 2016 for 0.5 mg and 0.3 mg specifications.⁵⁹ Through simplifying the process of drawing up medication from vials, use of prefilled syringe was shown to effectively reduce the risk of contamination and subsequent postinjection EO regardless of the anti-VEGF medication inside.^56,57 Ophthalmic use of bevacizumab was in the scope of off-label application; as preparation requires fractioning from the original vial into multiple injections, bacterial burden may accumulate by multiple punctures of the rubber cap of the vial.^7,39,60 In the US, bevacizumab syringes are commonly prepared by compounding pharmacies that have to comply with the US Pharmacopeia.^26,56 Bevacizumab was repackaged at compounding pharmacies in all of our studies in the office setting with available information on drug preparation, whereas 3 of 5 studies in the OR reported multiple withdrawals from a single vial by the treating physician. The relatively higher EO rates after bevacizumab administration in the OR may be a result of suboptimal aseptic rules followed during drug preparation in a setting other than the standard compounding pharmacy.¹² Owning to the sterile drug prepackaging procedures in a controlled environment, Bavinger et al⁵⁶ found that among 1 095 305 IVIs, compared with prefilled bevacizumab syringes at compounding pharmacies, office-filled ranibizumab and aflibercept together increased postinjection EO (odds ratio, 1.29; P = .02).

From our analysis and previous studies,⁶¹ injection setting did not appear to be an influential factor of postinjection EO. In a study of 14 001 anti-VEGF IVIs delivered in a procedure room dedicated for IVI only and with strict aseptic rules, serious ocular AEs, including EO, traumatic cataract, and RD, were observed only in 1 (0.01%), 3 (0.02%), and 1 (0.01%) cases, respectively.⁶² Therefore, we propose that a well-established IVI protocol together with sterile environment and aseptic procedure preparation are key to the safe delivery of anti-VEGF IVIs. Based on our sensitivity analysis in the OR setting, overall and culture-positive EO rates together with that of the bevacizumab subgroup were significantly higher in resource-limited countries than higher-income ones, yet no difference was found in the ranibizumab subgroup (eFigures 1 to 5 in the Supplement). However, we cannot determine from the data analyzed whether office-based injections in low-income regions have any greater rate of EO. The fact that there is a strong regional difference in setting, especially among OR-based studies where a significant heterogeneity of the data are observed, is a large limiting factor in making accurate comparisons.

The pooled rates of other ocular and systematic AEs were rare (1 or more per 10 000 to less than 1 per 1000) or very rare (less than 1 per 10 000) in both settings. Incorrect injection technique is a primary cause of posterior vitreous detachment; our findings were consistent with the previous report of a low rate (0 to 0.67%) of RD and RT following anti-VEGF IVI.⁶³ In a meta-analysis of 8 RCTs evaluating the systemic safety of intravitreal anti-VEGF agents for treatment of retinal vein occlusion, anti-VEGF drugs did not increase the risks of cardiovascular events compared with placebo/retinal photocoagulation.¹⁰

Limitations

There are several limitations to our study. First, given that EO rates are very low, findings may not be conclusive based on the current sample size. Second, information bias is possible, as the included studies almost exclusively focused on EO; other safety outcomes were rarely reported. Third, most studies are retrospective case series, which provides lower levels of evidence than RCTs and prospective cohorts. However, as the rates of EO were very low, prospective trials are not feasible and may lack clinical relevance even in the presence of statistical significance.⁴⁷ Fourth, the significant heterogeneity of data from OR-based injections may limit the strength of our findings. Fifth, since only 2 studies are directly comparative, our analysis was largely restricted to indirect comparison.

Conclusions

In conclusion, the rate of clinically suspected or culture-positive EO following anti-VEGF IVIs was low whether the procedure was performed in the office or OR. Bacterial spectrum can differ between the 2 settings. In resource-limited regions, we were unable to identify evidence that in-office procedures would lead to more EO than injections in the OR. Using antisepsis from drug packaging and loading the syringes to the completion of IVIs in a well-controlled, clean environment may contribute to prophylaxis of postinjection EO.

Supplement.

eMethods. Search strategies.

eTable 1. Baseline characteristics of the included studies.

eTable 2. Methodological qualities of the included studies.

eTable 3. Characteristics of the injection procedure and setting.

eFigure 1. Sensitivity analysis of EO rate in the office setting.

eFigure 2. Overall rate of EO following intravitreal anti-VEGF injections in the OR setting by country development status.

eFigure 3. Rate of EO following intravitreal ranibizumab injections in the OR setting by country development status.

eFigure 4. Rate of EO following intravitreal bevacizumab injections in the OR setting by country development status.

eFigure 5. Rate of culture-positive EO following intravitreal anti-VEGF injections in the OR setting by country development status.

eFigure 6. Egger test results.

Click here for additional data file.^{(639.7KB, pdf)}

References

1.Sampat KM, Garg SJ. Complications of intravitreal injections. Curr Opin Ophthalmol. 2010;21(3):178-183. doi: 10.1097/ICU.0b013e328338679a [DOI] [PubMed] [Google Scholar]
2.Busch C, Iglicki M, Okada M, et al. ; International Retina Group (IRG) . Causative pathogens of endophthalmitis after intravitreal anti-VEGF injection: an international multicenter study. Ophthalmologica. 2019;241(4):211-219. doi: 10.1159/000496942 [DOI] [PubMed] [Google Scholar]
3.Sheu SJ. Endophthalmitis. Korean J Ophthalmol. 2017;31(4):283-289. doi: 10.3341/kjo.2017.0036 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Kiss S, Dugel PU, Khanani AM, et al. Endophthalmitis rates among patients receiving intravitreal anti-VEGF injections: a USA claims analysis. Clin Ophthalmol. 2018;12:1625-1635. doi: 10.2147/OPTH.S169143 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Durand ML. Bacterial and fungal endophthalmitis. Clin Microbiol Rev. 2017;30(3):597-613. doi: 10.1128/CMR.00113-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Zhang Y, Liu ZL, Zhang H. Observation of intravitreal injections of ranibizumab for myopic choroidal neovascularization in Chinese patients. Int Eye Sci. 2015;15(3):381-385. doi: 10.3980/j.issn.1672-5123.2015.3.01 [DOI] [Google Scholar]
7.Mishra C, Lalitha P, Rameshkumar G, et al. Incidence of endophthalmitis after intravitreal injections: risk factors, microbiology profile, and clinical outcomes. Ocul Immunol Inflamm. 2018;26(4):559-568. doi: 10.1080/09273948.2018.1430238 [DOI] [PubMed] [Google Scholar]
8.Tabandeh H, Boscia F, Sborgia A, et al. Endophthalmitis associated with intravitreal injections: office-based setting and operating room setting. Retina. 2014;34(1):18-23. doi: 10.1097/IAE.0000000000000008 [DOI] [PubMed] [Google Scholar]
9.Patel SN, Gangaputra S, Sternberg P Jr, Kim SJ. Prophylaxis measures for postinjection endophthalmitis. Surv Ophthalmol. 2020;65(4):408-420. doi: 10.1016/j.survophthal.2019.12.005 [DOI] [PubMed] [Google Scholar]
10.Merani R, Hunyor AP. Endophthalmitis following intravitreal anti-vascular endothelial growth factor (VEGF) injection: a comprehensive review. Int J Retina Vitreous. 2015;1:9. doi: 10.1186/s40942-015-0010-y [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Sigford DK, Reddy S, Mollineaux C, Schaal S. Global reported endophthalmitis risk following intravitreal injections of anti-VEGF: a literature review and analysis. Clin Ophthalmol. 2015;9:773-781. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Zarei M, Karkhaneh R, Mahmoudzadeh R, et al. Incidence and risk factors for post-intravitreal injection endophthalmitis in a modified operating room setting. Ocul Immunol Inflamm. 2019;27(8):1314-1321. doi: 10.1080/09273948.2018.1526306 [DOI] [PubMed] [Google Scholar]
13.Abell RG, Kerr NM, Allen P, Vote BJ. Intravitreal injections: is there benefit for a theatre setting? Br J Ophthalmol. 2012;96(12):1474-1478. doi: 10.1136/bjophthalmol-2012-302030 [DOI] [PubMed] [Google Scholar]
14.Stroup DF, Berlin JA, Morton SC, et al. ; Meta-analysis Of Observational Studies in Epidemiology (MOOSE) Group . Meta-analysis of observational studies in epidemiology: a proposal for reporting. JAMA. 2000;283(15):2008-2012. doi: 10.1001/jama.283.15.2008 [DOI] [PubMed] [Google Scholar]
15.National Institute for Clinical Excellence . Appendix 4: quality assessment for case series. Accessed July 17, 2021. https://www.nice.org.uk/guidance/cg3/documents/appendix-4-quality-of-case-series-form2
16.Wells G, Shea B, O’Connell D, et al. Newcastle-Ottawa Quality Assessment Scale: case control studies. Accessed February 8, 2021. http://www.ohri.ca/programs/clinical_epidemiology/nosgen.pdf
17.Deeks JJ, Higgins JPT, Altman DG; Cochrane Statistical Methods Group . Chapter 10: analysing data and undertaking meta-analyses. Accessed February 8, 2021. https://training.cochrane.org/handbook/current/chapter-10
18.McCannel CA. Meta-analysis of endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor agents: causative organisms and possible prevention strategies. Retina. 2011;31(4):654-661. doi: 10.1097/IAE.0b013e31820a67e4 [DOI] [PubMed] [Google Scholar]
19.Lin L, Chu H. Meta-analysis of proportions using generalized linear mixed models. Epidemiology. 2020;31(5):713-717. doi: 10.1097/EDE.0000000000001232 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Schwarzer G, Chemaitelly H, Abu-Raddad LJ, Rücker G. Seriously misleading results using inverse of Freeman-Tukey double arcsine transformation in meta-analysis of single proportions. Res Synth Methods. 2019;10(3):476-483. doi: 10.1002/jrsm.1348 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.The World Bank . World Bank country and lending groups. Accessed February 8, 2021. https://datahelpdesk.worldbank.org/knowledgebase/articles/906519
22.Borkar DS, Obeid A, Su DC, et al. ; Wills Post Injection Endophthalmitis (PIE) Study Group . Endophthalmitis rates after bilateral same-day intravitreal anti-vascular endothelial growth factor injections. Am J Ophthalmol. 2018;194:1-6. doi: 10.1016/j.ajo.2018.06.022 [DOI] [PubMed] [Google Scholar]
23.Juncal VR, Francisconi CLM, Altomare F, et al. Same-day bilateral intravitreal anti-vascular endothelial growth factor injections: experience of a large Canadian retina center. Ophthalmologica. 2019;242(1):1-7. doi: 10.1159/000499115 [DOI] [PubMed] [Google Scholar]
24.Goldberg RA, Shah CP, Wiegand TW, Heier JS. Noninfectious inflammation after intravitreal injection of aflibercept: clinical characteristics and visual outcomes. Am J Ophthalmol. 2014;158(4):733-737.e1. doi: 10.1016/j.ajo.2014.06.019 [DOI] [PubMed] [Google Scholar]
25.Fintak DR, Shah GK, Blinder KJ, et al. Incidence of endophthalmitis related to intravitreal injection of bevacizumab and ranibizumab. Retina. 2008;28(10):1395-1399. doi: 10.1097/IAE.0b013e3181884fd2 [DOI] [PubMed] [Google Scholar]
26.Pilli S, Kotsolis A, Spaide RF, et al. Endophthalmitis associated with intravitreal anti-vascular endothelial growth factor therapy injections in an office setting. Am J Ophthalmol. 2008;145(5):879-882. doi: 10.1016/j.ajo.2007.12.036 [DOI] [PubMed] [Google Scholar]
27.Frenkel RE, Haji SA, La M, Frenkel MP, Reyes A. A protocol for the retina surgeon’s safe initial intravitreal injections. Clin Ophthalmol. 2010;4:1279-1285. doi: 10.2147/OPTH.S12846 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Haider MA, Imtiaz U, Javed F, Haider Z. Incidence of acute endophthalmitis after office based intravitreal bevacizumab injection. J Pak Med Assoc. 2017;67(12):1917-1919. [PubMed] [Google Scholar]
29.Rayess N, Rahimy E, Storey P, et al. Postinjection endophthalmitis rates and characteristics following intravitreal bevacizumab, ranibizumab, and aflibercept. Am J Ophthalmol. 2016;165:88-93. doi: 10.1016/j.ajo.2016.02.028 [DOI] [PubMed] [Google Scholar]
30.Storey P, Dollin M, Pitcher J, et al. ; Post-Injection Endophthalmitis Study Team . The role of topical antibiotic prophylaxis to prevent endophthalmitis after intravitreal injection. Ophthalmology. 2014;121(1):283-289. doi: 10.1016/j.ophtha.2013.08.037 [DOI] [PubMed] [Google Scholar]
31.Moshfeghi AA, Rosenfeld PJ, Flynn HW Jr, et al. Endophthalmitis after intravitreal vascular [corrected] endothelial growth factor antagonists: a six-year experience at a university referral center. Retina. 2011;31(4):662-668. doi: 10.1097/IAE.0b013e31821067c4 [DOI] [PubMed] [Google Scholar]
32.Chaudhary KM, Romero JM, Ezon I, Fastenberg DM, Deramo VA. Pars plana vitrectomy in the management of patients diagnosed with endophthalmitis following intravitreal anti-vascular endothelial growth factor injection. Retina. 2013;33(7):1407-1416. doi: 10.1097/IAE.0b013e3182807659 [DOI] [PubMed] [Google Scholar]
33.Shimada H, Hattori T, Mori R, Nakashizuka H, Fujita K, Yuzawa M. Minimizing the endophthalmitis rate following intravitreal injections using 0.25% povidone-iodine irrigation and surgical mask. Graefes Arch Clin Exp Ophthalmol. 2013;251(8):1885-1890. doi: 10.1007/s00417-013-2274-y [DOI] [PubMed] [Google Scholar]
34.Cheung CS, Wong AW, Lui A, Kertes PJ, Devenyi RG, Lam WC. Incidence of endophthalmitis and use of antibiotic prophylaxis after intravitreal injections. Ophthalmology. 2012;119(8):1609-1614. doi: 10.1016/j.ophtha.2012.02.014 [DOI] [PubMed] [Google Scholar]
35.Fineman MS, Hsu J, Spirn MJ, Kaiser RS. Bimanual assisted eyelid retraction technique for intravitreal injections. Retina. 2013;33(9):1968-1970. doi: 10.1097/IAE.0b013e318287da92 [DOI] [PubMed] [Google Scholar]
36.Englander M, Chen TC, Paschalis EI, Miller JW, Kim IK. Intravitreal injections at the Massachusetts Eye and Ear Infirmary: analysis of treatment indications and postinjection endophthalmitis rates. Br J Ophthalmol. 2013;97(4):460-465. doi: 10.1136/bjophthalmol-2012-302435 [DOI] [PubMed] [Google Scholar]
37.Jan S, Nazim M, Karim S, Hussain Z. Intravitreal bevacizumab: indications and complications. J Ayub Med Coll Abbottabad. 2016;28(2):364-368. [PubMed] [Google Scholar]
38.Freiberg FJ, Brynskov T, Munk MR, et al. Low endophthalmitis rates after intravitreal anti-vascular endothelial growth factor injections in an operation room: a retrospective multicenter study. Retina. 2017;37(12):2341-2346. doi: 10.1097/IAE.0000000000001488 [DOI] [PubMed] [Google Scholar]
39.Wani VB, Al-Kandari J, Sabti K, et al. Incidence of endophthalmitis after intravitreal bevacizumab using aliquots prepared on-site in 2 operating rooms in Kuwait. Middle East Afr J Ophthalmol. 2016;23(1):64-70. doi: 10.4103/0974-9233.171784 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Hasler PW, Bloch SB, Villumsen J, Fuchs J, Lund-Andersen H, Larsen M. Safety study of 38,503 intravitreal ranibizumab injections performed mainly by physicians in training and nurses in a hospital setting. Acta Ophthalmol. 2015;93(2):122-125. doi: 10.1111/aos.12589 [DOI] [PubMed] [Google Scholar]
41.Casparis H, Wolfensberger TJ, Becker M, et al. Incidence of presumed endophthalmitis after intravitreal injection performed in the operating room: a retrospective multicenter study. Retina. 2014;34(1):12-17. doi: 10.1097/IAE.0b013e31829f74b0 [DOI] [PubMed] [Google Scholar]
42.Ou YL, Zhou XP, Xie LL, et al. Anti-VEGF combined with retinal laser in treatment of retinal vein occlusion with macular edema. Int Eye Sci. 2019;19(7):1162-1165. doi: 10.3980/j.issn.1672-5123.2019.7.18 [DOI] [Google Scholar]
43.Zafar S, Hamid A, Bin Mahmood SU, Burq MA, Maqsood N. Incidence of endophthalmitis after intravitreal injections at a tertiary care hospital. Can J Ophthalmol. 2018;53(2):94-97. doi: 10.1016/j.jcjo.2017.07.027 [DOI] [PubMed] [Google Scholar]
44.Istek S, Bursali Ö, Alagöz G. Intravitreal bevacizumab injection in the treatment of macular edema secondary to branch retinal vein occlusion. Int Eye Sci. 2014;14(6):979-985. doi: 10.3980/j.issn.1672-5123.2014.06.01 [DOI] [Google Scholar]
45.Brynskov T, Kemp H, Sørensen TL. No cases of endophthalmitis after 20,293 intravitreal injections in an operating room setting. Retina. 2014;34(5):951-957. doi: 10.1097/IAE.0000000000000071 [DOI] [PubMed] [Google Scholar]
46.Nentwich MM, Yactayo-Miranda Y, Schwarzbach F, Wolf A, Kampik A, Mino de Kaspar H. Endophthalmitis after intravitreal injection: decreasing incidence and clinical outcome-8-year results from a tertiary ophthalmic referral center. Retina. 2014;34(5):943-950. doi: 10.1097/IAE.0000000000000011 [DOI] [PubMed] [Google Scholar]
47.Lau PE, Jenkins KS, Layton CJ. Current evidence for the prevention of endophthalmitis in anti-VEGF intravitreal injections. J Ophthalmol. 2018;2018:8567912. doi: 10.1155/2018/8567912 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Xing L, Dorrepaal SJ, Gale J. Survey of intravitreal injection techniques and treatment protocols among retina specialists in Canada. Can J Ophthalmol. 2014;49(3):261-266. doi: 10.1016/j.jcjo.2014.03.009 [DOI] [PubMed] [Google Scholar]
49.Samia-Aly E, Cassels-Brown A, Morris DS, Stancliffe R, Somner JE. A survey of UK practice patterns in the delivery of intravitreal injections. Ophthalmic Physiol Opt. 2015;35(4):450-454. doi: 10.1111/opo.12217 [DOI] [PubMed] [Google Scholar]
50.Haddock LJ, Ramsey DJ, Young LH. Complications of subspecialty ophthalmic care: endophthalmitis after intravitreal injections of anti-vascular endothelial growth factor medications. Semin Ophthalmol. 2014;29(5-6):257-262. doi: 10.3109/08820538.2014.959616 [DOI] [PubMed] [Google Scholar]
51.Chen E, Lin MY, Cox J, Brown DM. Endophthalmitis after intravitreal injection: the importance of viridans streptococci. Retina. 2011;31(8):1525-1533. doi: 10.1097/IAE.0b013e318221594a [DOI] [PubMed] [Google Scholar]
52.Green-Simms AE, Ekdawi NS, Bakri SJ. Survey of intravitreal injection techniques among retinal specialists in the United States. Am J Ophthalmol. 2011;151(2):329-332. doi: 10.1016/j.ajo.2010.08.039 [DOI] [PubMed] [Google Scholar]
53.Shah CP, Garg SJ, Vander JF, Brown GC, Kaiser RS, Haller JA; Post-Injection Endophthalmitis (PIE) Study Team . Outcomes and risk factors associated with endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor agents. Ophthalmology. 2011;118(10):2028-2034. doi: 10.1016/j.ophtha.2011.02.034 [DOI] [PubMed] [Google Scholar]
54.Simonett JM, Igelman A, Taylor SC, et al. Culture-proven endophthalmitis after intravitreal injection: a 10-year analysis. Ophthalmic Surg Lasers Imaging Retina. 2019;50(1):33-38. doi: 10.3928/23258160-20181212-05 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Ramos MS, Xu LT, Singuri S, et al. Patient-reported complications after intravitreal injection and their predictive factors. Ophthalmol Retina. 2021;5(7):625-632. doi: 10.1016/j.oret.2020.09.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Bavinger JC, Yu Y, VanderBeek BL. Comparative risk of endophthalmitis after intravitreal injection with bevacizumab, aflibercept, and ranibizumab. Retina. 2019;39(10):2004-2011. doi: 10.1097/IAE.0000000000002351 [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Baudin F, Benzenine E, Mariet AS, et al. Association of acute endophthalmitis with intravitreal injections of corticosteroids or anti-vascular growth factor agents in a nationwide study in France. JAMA Ophthalmol. 2018;136(12):1352-1358. doi: 10.1001/jamaophthalmol.2018.3939 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Hahn P, Kim JE, Stinnett S, et al. ; American Society of Retina Specialists Therapeutic Surveillance Committee . Aflibercept-related sterile inflammation. Ophthalmology. 2013;120(5):1100-1101.e5. doi: 10.1016/j.ophtha.2012.11.018 [DOI] [PubMed] [Google Scholar]
59.Schargus M, Frings A. Issues with intravitreal administration of anti-VEGF drugs. Clin Ophthalmol. 2020;14:897-904. doi: 10.2147/OPTH.S207978 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Sassalos TM, Paulus YM. Prefilled syringes for intravitreal drug delivery. Clin Ophthalmol. 2019;13:701-706. doi: 10.2147/OPTH.S169044 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Bande MF, Mansilla R, Pata MP, et al. Intravitreal injections of anti-VEGF agents and antibiotic prophylaxis for endophthalmitis: a systematic review and meta-analysis. Sci Rep. 2017;7(1):18088. doi: 10.1038/s41598-017-18412-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Xu Y, Tan CS. Safety and complications of intravitreal injections performed in an Asian population in Singapore. Int Ophthalmol. 2017;37(2):325-332. doi: 10.1007/s10792-016-0241-4 [DOI] [PubMed] [Google Scholar]
63.Falavarjani KG, Nguyen QD. Adverse events and complications associated with intravitreal injection of anti-VEGF agents: a review of literature. Eye (Lond). 2013;27(7):787-794. doi: 10.1038/eye.2013.107 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials