Antivascular endothelial growth factor (VEGF) agents have transformed the clinical management of sight-threatening retinal diseases such as age-related macular degeneration, diabetic maculopathy,1,2 branch and central retinal vein occlusion,3,4 and neovascular glaucoma.5,6 Since their introduction as intravitreal therapies, anti-VEGF agents have become the gold-standard treatment in these conditions. Although the safety profile of intravitreal injections (IVIs) is well established, there are only a handful of studies that focus on the possible impact of multiple previous IVI procedures on cataract surgery outcomes. Patients often require multiple injections as a result of suggested loading regimens, partially because of the relatively short intravitreal half-life of these medications.7
Current guidance on safe administration of IVIs states that the site of the injection should be 4.0 mm behind the limbus if phakic and 3.5 mm if pseudophakic. It is also recommended that patients are supine to prevent lens touch.8 The proximity of the needle to the lens may be associated with an increased risk of lenticular trauma either subclinical or through needle penetration, thus increasing the risk of posterior capsular rupture (PCR) during phacoemulsification surgery.9 Two landmark trials, ANCHOR2 and MARINA,10 reported only one case of lens trauma as a complication, but recent real-world studies have shown a higher proportion of PCR as a result of previous IVIs.9,11,12 Moreover, previous studies have reported an increased risk of PCR in patients who received IVIs before cataract surgery, in the absence of obvious damage to the posterior capsule (lens touch).9,13,14
The aim of this study is to further analyze whether previous intravitreal anti-VEGF injections are a significant risk factor for PCR during cataract surgery, despite the absence of signs of visible trauma, and whether there was a cumulative risk in eyes that received multiple injections.
This was a retrospective, single-center, comparative, noninterventional, cohort study approved by the Barts Health NHS Trust Clinical Effectiveness Board (project number 8565). This study was conducted in accordance with the tenets of the Declaration of Helsinki and the United Kingdom Data Protection Act. Anonymized data were extracted using the electronic healthcare record system, Medisoft (Medisoft Limited), from Whipps Cross University Hospital over the period of August 1, 2016, to January 1, 2018. Whipps Cross University Hospital is based in East London, the United Kingdom, and is the sole NHS provider of eyecare for a population of 650 000 people. The study period was chosen to coincide with a period immediately following a department-wide recognition of a possible link between IVI and PCR. As such, all patients with a history of IVIs underwent careful slitlamp examination of the posterior capsule, with any adverse findings documented.
Demographic data, including age, sex, ethnicity, and diabetic status, as well as the indication for treatment, drug name, and number of previous intravitreal anti-VEGF injections were collected. Surgeon grade and intraoperative complications were included for the univariate analysis. The main inclusion criterion was phacoemulsification cataract surgery during the study period. Exclusion criteria were patients younger than 18 years, nonphacoemulsification cataract extraction techniques, and cataract surgery combined with any other intraocular procedures including IVIs at the time of cataract surgery or combined minimally invasive glaucoma surgery. For patients undergoing cataract surgery on both eyes, only 1 eye was included in the study cohort. The study cohort was further subdivided into 2 groups, those that had received at least one IVI before cataract surgery and those that had not.
The primary outcome measure was PCR during phacoemulsification cataract surgery as defined by the Royal College of Ophthalmologists' National Ophthalmology Database audit of cataract surgery.15 This included events such as zonular rupture, loss of nucleus into the vitreous, intraocular lens into the vitreous, and nonspecified vitreous loss.
All datasets collected were entered into a database created using Microsoft Excel 2010 (Microsoft Corp.). The statistical analysis was conducted using IBM SPSS Statistics for Windows software (version 25.0, IBM Corp.). Any eyes with preoperative documentation of a visible breach of the posterior capsule were excluded from the final analysis. For continuous variables, the normality of data distribution was determined using the Shapiro-Wilks test, in which normality was defined as a P value of .05 or greater. For normally distributed data, comparison between subgroups was performed using the t test; for non-normally distributed data, the Wilcoxon matched-pair signed-rank and Mann-Whitney U tests were used for matched and unmatched data, respectively. Categorical variables were compared using the χ2 test. The univariate logistic regression analysis was performed to assess the association between the number of previous intravitreal anti-VEGF injections and the occurrence of PCR.
During the study period, 4047 eyes from 4047 patients were included for the analysis. The background demographic information is described in Table 1. IVIs were provided by 15 trained clinicians, including 3 retinal consultants, 4 postresidency retinal fellows, and 8 residents. Surgery was performed by 10 consultants and 12 junior surgeons of differing grades. The overall rate of PCR was 2.08% (84/4047), and visual acuity was significantly improved from 0.64 ± 0.57 to 0.25 ± 0.32 logarithm of the minimum angle of resolution (P < .0001). One third of the surgery was performed on eyes of black or Asian ethnicity, and over a third had a history of diabetes mellitus.
One hundred eight eyes (2.7%) had received previous anti-VEGF injections. Within this group, the mean number of injections received was 10.4 ± 8.1 (Figure 1).
Three eyes in the IVI subgroup were noted to have visible preoperative lenticular damage to the PC. The rate of PCR in the IVI group including the 3 eyes was 9.26% (10/108). The PCR rate in the IVI group excluding these 3 eyes was 6.67% (7/105). In both groups, the rates of PCR were significantly higher than those in the non-IVI group (1.88%; odds ratio [OR], 4.93; P < .0001 and OR, 3.55; P = .0047, respectively) (Table 2).
The intraoperative stage at which PCR occurred in the 7 cases and at which dropped nucleus occurred in the 2 cases are reported in Table 3. Immediate anterior vitrectomy was performed in all cases; in the 2 cases of dropped nucleus, a same-day referral to a vitreoretinal specialist was made and subsequent pars plana vitrectomy was performed.
The univariate logistic regression analysis of the number of previous injections as a risk factor for PCR showed a dose-dependent increase of 8.6% relative risk per injection (OR, 1.086; 95% CI, 1.040-1.135; P = .0002). The PCR rate was higher in those who received more than 10 injections compared with those who received 10 injections or less (14.3% vs 6.1%, P = .18).
Within this cohort, patients' ethnicity and diabetic status were not found to be statistically significant risk factors for PCR. Similarly, the grade of the operating surgeon also did not have a statistically significant effect. Further subgroup analysis based on the type of the anti-VEGF agent was not performed because of a large proportion of this cohort having received multiple agents during their treatment.
The results from 4047 eyes undergoing phacoemulsification cataract surgery showed that previous intravitreal therapy is associated with a clinically and statistically significant increased risk of PCR even in the absence of visible PC trauma. Although the exact mechanism is unclear, it has been speculated that there may be local scleral deformation during the administration of the anti-VEGF agent11 causing inadvertent zonular trauma, but there may also be biochemical factors of which we are not aware. Previous studies have shown that administration of certain classes of intravitreal agents has been associated with cataract formation and zonular dehiscence, but once again the mechanism is unclear.16,17
Cases of unidentified posterior capsule trauma have the potential to cause significant complications during surgery, hence the importance of preoperative identification. Of note, in our study cohort, PCR did occur in eyes in which no evidence of posterior capsule trauma was specifically recorded in the notes. Studies have shown that optical coherence tomography lens reconstruction can provide accurate and quantitative estimates of lens volume, size, and plane.18 There may be room to use this technology in higher risk groups (>10 IVIs) to aid in the preoperative identification of PC trauma.
The patient demographics for the anti-VEGF cohort showed a higher proportion of white, British, female patients with a higher incidence of diabetes (including diabetic retinopathy) and age-related macular degeneration. However, these underlying factors did not account for the large difference in PCR rates in between the 2 groups.
The likelihood of PCR was approximately 5 times higher (9.26% vs 1.88%) in eyes with previous anti-VEGF IVI treatments. Even after excluding the 3 eyes with preoperatively identified PC trauma, the likelihood of PCR remained more than 3 times higher (6.67% vs 1.81%). The results were statistically significant in both analyses.
This study also found a dose-dependent increase of 8.6% relative risk per injection (OR, 1.086) in patients, which was similar to a study by Lee et al.11 (OR, 1.039). In addition to this, they reported 10 or more previous injections were associated with a 2.59 times higher likelihood of PCR. Our subgroup analysis showed patients who had received more than 10 injections had a comparable PCR rate of 14.3% (P = .18), which was 2.36 times higher than patients who had received 10 or fewer injections.
The recent Royal College of Ophthalmologists' National Ophthalmology Database audit of cataract surgery found the PCR rate to be 1.92%.15 The risk of PCR after previous IVI treatment (OR, 4.93) is comparable to no fundal view/vitreous opacities (OR, 4.67) or brunescent cataract (OR, 4.19) and is an important consideration when discussing surgery with patients. Informed consent should include a discussion of the much higher risk of PCR with previous anti-VEGF injections. Surgical teams can be alerted to higher risk situations, and this allows for preoperative planning to use risk reduction strategies and risk avoidance by junior trainees. It may be prudent to list these patients, especially those with more than 10 injections, on a day with a vitreoretinal cover.
We acknowledge that this study was conducted in a single center and was limited by its retrospective nature. Furthermore, despite a cohort of more than 4000 patients, the IVI group consisted of 108 patients and therefore formed a relatively small proportion of the overall dataset. As such, subgroup analyses were not possible because they would be of insufficient statistical power. It would be very informative to expand this study to a pooled, multicenter database to further our understanding of this issue.
In summary, the results of this study have shown that previous IVI anti-VEGF administration is a significant risk factor for PCR during phacoemulsification surgery and that this risk is dose-dependent and higher for eyes that have received more than 10 injections. PCR occurred during different surgical steps with no obvious cause.
WHAT WAS KNOWN
- There is some evidence of the risk of posterior capsular rupture (PCR) after previous intravitreal antivascular endothelial growth factor (VEGF) treatment. It is not recognized as a significant risk factor in the National Ophthalmology Database audit.
WHAT THIS PAPER ADDS
- There is a significant risk of PCR with previous intravitreal anti-VEGF injections.
- There is a cumulative risk with a clinically significant increase in risk with patients who have had more than 10 injections before surgery. The risk is comparable to high risk factors such as poor fundal view or brunescent cataracts.
- It is worth considering listing patients with high numbers of previous injections with a suitable vitreoretinal cover.
1. Heier JS, Brown DM, Chong V, Korobelnik J-F, Kaiser PK, Nguyen QD, Kirchhof B, Ho A, Ogura Y, Yancopoulos GD, Stahl N, Vitti R, Berliner AJ, Soo Y, Anderesi M, Groetzbach G, Sommerauer B, Sandbrink R, Simader C, Schmidt-Erfurth U; VIEW 1 and VIEW 2 Study Groups. Intravitreal aflibercept (VEGF trap-eye) in wet age-related macular degeneration. Ophthalmology 2012;119:2537–2548
2. Brown DM, Kaiser PK, Michels M, Soubrane G, Heier JS, Kim RY, Sy JP, Schneider S; ANCHOR Study Group. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med 2006;355:1432–1444
3. Jampol LM, Glassman AR, Bressler NM. Comparative effectiveness trial for diabetic macular edema. JAMA Ophthalmol 2015;133:983
4. Wells JA, Glassman AR, Ayala AR, Jampol LM, Bressler NM, Bressler SB, Brucker AJ, Ferris FL, Hampton GR, Jhaveri C, Melia M, Beck RW; Diabetic Retinopathy Clinical Research Network. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema: two-year results from a comparative effectiveness randomized clinical trial. Ophthalmology 2016;123:1351–1359
5. Schmidt-Erfurth U, Eldem B, Guymer R, Korobelnik JF, Schlingemann RO, Axer-Siegel R, Wiedemann P, Simader C, Gekkieva M, Weichselberger A; EXCITE Study Group. Efficacy and safety of monthly versus quarterly ranibizumab treatment in neovascular age-related macular degeneration: the EXCITE study. Ophthalmology 2011;118:831–839
6. Wolf S, Balciuniene VJ, Laganovska G, Menchini U, Ohno-Matsui K, Sharma T, Wong TY, Silva R, Pilz S, Gekkieva M; RADIANCE Study Group. RADIANCE: a randomized controlled study of ranibizumab in patients with choroidal neovascularization secondary to pathologic myopia. Ophthalmology 2014;121:682–692.e2
7. Avery RL, Castellarin AA, Steinle NC, Dhoot DS, Pieramici DJ, See R, Couvillion S, Nasir MA, Rabena MD, Maia M, Van Everen S, Le K, Hanley WD. Systemic pharmacokinetics and pharmacodynamics of intravitreal aflibercept, bevacizumab, and ranibizumab. Retina 2017;37:1847–1858
8. Shalchi Z, Okada M, Whiting C, Hamilton R. Risk of posterior capsule rupture during cataract surgery in eyes with previous intravitreal injections. Am J Ophthalmol 2017;177:77–80
9. Khalifa YM, Pantanelli SM. Quiescent posterior capsule trauma after intravitreal injection: implications for the cataract surgeon. J Cataract Refract Surg 2011;37:1364
10. Rosenfeld PJ, Brown DM, Heier JS, Boyer DS, Kaiser PK, Chung CY, Kim RY; MARINA Study Group. Ranibizumab for neovascular age-related macular degeneration. New Engl J Med Med 2006;355:1419–1431
11. Lee AY, Day AC, Egan C, Bailey C, Johnston RL, Tsaloumas MD, Denniston AK, Tufail A. Previous intravitreal therapy is associated with increased risk of posterior capsule rupture during cataract surgery. Ophthalmology 2016;123:1252–1256
12. Saeed MU, Prasad S. Management of cataract caused by inadvertent capsule penetration during intravitreal injection of ranibizumab. J Cataract Refract Surg 2009;35:1857–1859
13. Falavarjani KG, Nguyen QD. Adverse events and complications associated with intravitreal injection of anti-VEGF agents: a review of literature. Eye 2013;27:787–794
14. Jonas JB, Spandau UH, Schlichtenbrede F. Short-term complications of intravitreal injections of triamcinolone and bevacizumab. Eye (Lond) 2008;22:590–591
15. Henry P, Donachie J, Sparrow JM. Year 3 annual report-the second prospective report of the National Ophthalmology Database Audit National Ophthalmology Database Audit 2 NOD audit third annual report-second prospective audit year report [internet]. 2018. Available at: https://www.nodaudit.org.uk/u/docs/20/avusuryktz/NOD
Audit Annual Report 2018.pdf. Accessed January 22, 2019.
16. Hurley B. Cataract formation and other complications of intravitreal triamcinolone for macular edema. Evidence-Based Ophthalmol 2006;141(4):629–637
17. Keller J, Haynes RJ. Zonular dehiscence at the time of combined vitrectomy and cataract surgery after intravitreal ocriplasmin injection. JAMA Ophthalmol 2015;133:1091–1092
18. Martinez-Enriquez E, Sun M, Velasco-Ocana M, Birkenfeld J, Pérez-Merino P, Marcos S. Optical coherence tomography based estimates of crystalline lens volume, equatorial diameter, and plane position. Investig Ophthalmol Vis Sci 2016;57:OCT600–OCT610