Tetsuya Muto, Shigeki Machida, Shinichiro Imaizumi

1Department of Ophthalmology, Dokkyo Medical University Saitama Medical Center, Koshigaya 343-8555, Japan

2Imaizumi Eye Hospital, Koriyama 963-8877, Japan

3Fukushima Medical University, Fukushima 960-1295, Japan

Abstract

● KEYWORDS: aflibercept; intraocular pressure; vitreous reflux; intravitreal injection

INTRODUCTION

Nowadays, intravitreal aflibercept (Eylea?; Regeneron,Tarrytown, NY, USA) injection (IVAI) is extensively used for diabetic macular edema (DME)[1], neovascular agerelated macular degeneration (nAMD)[2], cystoid macular edema (CME) secondary to retinal vein occlusion (RVO)[3-4],and myopic choroidal neovascularization (CNV)[5]treatment in Japan.Intravitreal injection of anti-vascular endothelial growth factor (VEGF) antibody including aflibercept, causes animmediate increase of the intraocular pressure (IOP)[6-8].The immediate IOP increase is one of the risk factors for irreversible optic nerve head change[7].Therefore, an immediate IOP increase should be avoided as much as possible.

Vitreous reflux (VR) is frequently observed after intravitreal anti-VEGF antibody injection.VR ameliorates the immediate IOP increase[9-11].Uyaret al[9]reported that VR rate decreased as the total number of intravitreal ranibizumab injection increased.Accordingly, the immediate IOP increased in proportion to the total number of intravitreal ranibizumab injections.These phenomena might be related to scleral thinning because repeated intravitreal anti-VEGF antibody injections caused scleral thinning around the injection sites[12].In contrast, Demirelet al[13]reported that the total number of intravitreal ranibizumab injections was not correlated with immediate IOP increase.The correlation between the total number of IVAIs and VR rate, and immediate IOP increase is still unclear.Furthermore, the correlation between background diseases (CME secondary to RVO, DME, nAMD, and myopic CNV) and immediate IOP increase after IVAI also remains to be investigated.Therefore, the total number of IVAIs and background disease should be integrated in case of discussion of immediate IOP increase and VR after injection.Lens status has also been reported to affect immediate IOP increase or VR rate after intravitreal anti-VEGF antibody injection[13].Thus, we shall examine whether the background diseases and numbers of IVAIs affect VR rate and immediate IOP increase in phakic eyes.

SUBJECTS AND METHODS

Ethical ApprovalThe study adhered to the tenets of the Declaration of Helsinki, and its protocol was approved by the Institutional Review Board at Dokkyo Medical University Saitama Medical Center (approval number 21066).Informed consent was waived due to the retrospective nature of the study.

A total of 275 eyes of 275 patients with either CME secondary to RVO, DME, nAMD, or myopic CNV who visited Dokkyo Medical University Saitama Medical Center between February 2017 and December 2022 were studied.All eyes were phakic.All patients underwent the first or multiple IVAIs (2 mg/0.05 mL)performed by one of the authors (Muto T).Overall, 54 of 275 cases of multiple IVAI had a history of ＞10 times injection(13.7±4.9 times) for nAMD.The IOPs were measured before and immediately after IVAIs using an ICare PRO?(Tiolat,Helsinki, Finland).The immediate postinjection IOPs were consistently measured within 1min after the IVAIs.The amount of VR was evaluated by measuring the diameter of the broadest conjunctival formation with a ruler.Patients were allocated to one of the three groups according to their VR status: group 1, patients had no VR; group 2, patients had a conjunctival bleb diameter of ＜3 mm; and group 3,patients had a conjunctival bleb diameter of ＞3 mm[9,14].The axial length, anterior chamber depth and lens thickness of all patients were measured using an IOLMaster 700 (Carl Zeiss Meditec AG, Jena, Germany).Patients who used antiglaucoma eye solutions, had a history of intravitreal injection of other drugs (ranibizumab, bevacizumab, and triamcinolone acetonide) or subtenon injection of triamcinolone acetonide or vitrectomy, or had undergone cataract or glaucoma surgery were excluded from this study.Phakic patients who had a history of retinal photocoagulation were included in this study.None of the patients had used a Honan IOP reducer before intravitreal injection.

Conjunctival anesthesia was topically induced by instillation of 4% lidocaine (Xylocaine?solution 4%; Aspen Japan,Tokyo, Japan).The eyelids and ocular surface were disinfected with 0.027% iodine (PA●IODO?ophthalmic and eye washing solution; Nitten Pharmaceutical Co., Ltd., Nagoya, Japan).In the operating room, aflibercept was injected into the superotemporal quadrantviathe pars plana and into the vitreous cavity 3-4 mm posterior to the limbus using a 32-gauge needle (Dentronics 32 G?; Dentronics, Tokyo,Japan).Postinjection light perception was assessed.No eyes received anterior chamber paracenteses.Topical 0.5%levofloxacin solution (0.5% Cravit?ophthalmic solution;Santen Pharmaceutical, Osaka, Japan) was instilled four times daily for 3d before and after each intravitreal injection.

Statistical AnalysisAll data are expressed as mean±standard deviation (SD).One-way repeated-measures analysis of variance (ANOVA) was used to determine the statistical significance of patient’s age, immediate preinjection IOP,immediate postinjection IOP, anterior chamber depth, lens thickness, and axial length in four background diseases.Additionally, Tukey tests were conducted after the ANOVA as post-hoc tests.Chi-square test was used to determine the statistical significance of sex, right or left eye, and VR rate.Pearson’s correlation coefficient was used to evaluate the correlation of age with immediate postinjection IOP, and VR rate as well as correlation of axial length with immediate postinjection IOP, and VR rate.The unpairedt-test was used to determine the statistical significance of patient’s age,immediate preinjection IOP, immediate postinjection IOP,anterior chamber depth, lens thickness, and axial length between patients with first-time IVAI and multiple IVAIs.These analyses were conducted using the StatMate version V for Macintosh (ATMS, Tokyo, Japan).Differences with aP-value ＜0.05 were considered statistically significant.

RESULTS

Patient characteristics in four background diseases are shown in Table 1.There were no statistically significant differences in sex (P=0.35) and ratio of right to left eyes (P=0.21) among the patients.The mean age of patients with CME secondary to RVO was older than that in patients with DME (P＜0.01) and myopic CNV (P＜0.01).The mean age of patients with nAMD was older than that in patients with CME secondary to RVO(P＜0.001), DME (P＜0.001) and myopic CNV (P＜0.001).The mean anterior chamber depth of myopic CNV was the deepest among background diseases (P＜0.01).Lens thickness of myopic CNV was the thinnest among background diseases(P＜0.01).Lens thickness of CME secondary to RVO was thinner than that in patients with nAMD (P＜0.05).Axial length of myopic CNV was the longest among background diseases(P＜0.001) and axial length of nAMD was the shortest among background diseases (P＜0.05).

The immediate pre- and post-injection IOP and VR rate in the four diseases are shown in Figure 1.The immediate preinjection IOPs were 16.1±3.1 mm Hg in patients with CMEsecondary to RVO, 16.1±3.3 mm Hg in patients with DME,15.6±2.3 mm Hg in patients with nAMD, and 17.3±2.8 mm Hg in patients with myopic CNV (Figure 1A).The immediate postinjection IOPs were 51.7±12.3 mm Hg in patients with CME secondary to RVO, 53.0±13.7 mm Hg in patients with DME, 50.3±13.3 mm Hg in patients with nAMD, and 43.1±19.7 mm Hg in patients with myopic CNV (Figure 1B).Of 105 patients with CME secondary to RVO, 50, 28, and 27 patients were classified into groups 1, 2, and 3, respectively,according to the degree of VR.Of 35 patients with DME,22, 5, and 8 patients of the 35 patients were classified into groups 1, 2, and 3, respectively.Of 69 patients with nAMD,36, 21, and 12 patients were assigned to groups 1, 2, and 3,respectively.Similarly, of 12 patients with myopic CNV, 5, 4,and 3 patients were assigned to groups 1, 2, and 3, respectively(Figure 1C).There were no statistically significant amonggroup differences in immediate preinjection IOP (P=0.27),immediate postinjection IOP (P=0.16), and VR rate (P=0.50).Pearson’s correlation coefficient revealed no significant correlations between immediate postinjection IOP and age(CME secondary to RVO:P=0.23, DME:P=0.14, nAMD:P=0.68, and myopic CNV:P=0.48), between VR rate and age(CME secondary to RVO:P=0.53, DME:P=0.50, nAMD:P=0.24, and myopic CNV:P=0.99), between immediate postinjection IOP and axial length (CME secondary to RVO:P=0.074, DME:P=0.18, nAMD:P=0.44, and myopic CNV:P=0.11), or between VR rate and axial length (CME secondary to RVO:P=0.15, DME:P=0.42, nAMD:P=0.41, and myopic CNV:P=0.81) in patients with the four background diseases.The characteristics of patients with nAMD are shown in Table 2.There were no statistically significant differences in sex (P=0.54), mean age (P=0.31), ratio of right to left eyes(P=0.78), anterior chamber depth (P=0.79), lens thickness(P=0.58), and axial length (P=0.074) between the groups.

Table 1 Patient characteristics in the background diseases mean±SD

The immediate pre- and post-injection IOP and VR rate in patients with nAMD are shown in Figure 2.The immediate preinjection IOPs were 15.6±2.3 mm Hg in patients with first-time IVAI, and 15.9±2.7 in patients with multiple IVAIs (Figure 2A).The immediate postinjection IOPs were50.3±13.3 mm Hg in first-time IVAI, and 49.3±11.2 in multiple IVAIs (Figure 2B).Regarding VR, group 1, 2, and 3 consisted of 36, 21, and 12 patients of 69 patients with first-time IVAI,respectively.Of 54 patients with multiple IVAIs, groups 1, 2,and 3 consisted of 35, 10, and 9 patients, respectively (Figure 2C).There were no statistically significant differences between the groups in immediate preinjection IOP (P=0.58), immediate postinjection IOP (P=0.66), and VR rate (P=0.28) between the groups.

Table 2 Patient characteristics of first-time IVAI and multiple IVAIs in phakic patients with nAMD mean±SD

DISCUSSION

To the best of our knowledge, there have been no reports about the effect of background diseases on immediate IOP increase and VR rate after the first-time intravitreal injection.If the total numbers of intravitreal injections are different, deviation of immediate IOP increase and VR rates may occur.Hoanget al[15]reported that sustained IOP increase may be associated with numbers of intravitreal injections.To avoid deviation,the injection number was limited for the first-time only in the current research.Several reports did not limit injected agents with one type[16-18].Generally, 0.1 mL of triamcinolone acetonide and 0.05 mL of anti-VEGF agents are injected to the human eye.Characteristics, such as viscosity, may be different among injected agents even though the amount is the same.There were no statistically significant among-group differences in immediate IOP increase, and VR rate after first-time IVAI.However, there were statistically significant among-group differences in mean age, anterior chamber depth,lens thickness, and axial length.Uyaret al[9]reported that age was not correlated with VR rate in intravitreal ranibizumab injection and our results supported their study.Cacciamaniet al[19]and Gismondiet al[20]reported that axial length was negatively correlated with immediate IOP increase after intravitreal anti-VEGF antibody injection.However, other researchers detected no association between immediate IOP increase and axial length and our results supported their study[9,17,21].Although our results were different from their reports[19-20], they used bevacizumab[19]and ranibizumab[20], not aflibercept.Furthermore, Gismondiet al[20]did not mention about injection numbers in their report.It is difficult to simply compare the results.

Figure 1 Preinjection IOP, postinjection IOP, and VR rate among the four background diseases A: Preinjection IOP; There were no significant differences in immediate preinjection IOP (P=0.27) among the four background diseases.B: Postinjection IOP; There were no significant differences in immediate postinjection IOP (P=0.16) among the four background diseases.C: VR rate; There were no significant differences in VR rate (P=0.50) among the four background diseases.IOP: Intraocular pressure; RVO: Retinal vein occlusion; VR: Vitreous reflux; CME: Cystoid macular edema; DME: Diabetic macular edema; AMD: Age-related macular edema; CNV: Choroidal neovascularization.

Figure 2 Preinjection IOP, postinjection IOP, and VR rate of first-time IVAI and multiple IVAI in phakic patients with nAMD A: Preinjection IOP; There were no significant differences between first-time IVAI and multiple IVAIs in immediate preinjection IOP of nAMD (P=0.58).B:Postinjection IOP; There were no significant differences between first-time IVAI and multiple IVAIs in immediate postinjection IOP of nAMD(P=0.66).C: VR rate; There were no significant differences between first-time IVAI and multiple IVAIs in VR rate of nAMD (P=0.28).VR: Vitreous reflux; IOP: Intraocular pressure; nAMD: Neovascular age-related macular edema; IVAI: Intravitreal aflibercept injection.

Uyaret al[9]reported that the number of previous intravitreal ranibizumab injection was negatively correlated with VR rate.On the other hand, Lemoset al[22]reported that the number of previous intravitreal bevacizumab injections was not correlated with immediate IOP increase.The effect of the number of previous IVAIs on immediate IOP increase and VR rate was confirmed in nAMD of phakic eyes in the current study.The effect of the number of previous IVAIs does not need to be considered in case of IVAI for nAMD of phakic eyes from our results.Although the results of Uyaret al’s[9]report and ours were different, we thought the reason was character difference between ranibizumab and aflibercept.At least molecular weights were different from those of them.Although only nAMD of phakic eyes was evaluated in the current study, other background diseases in phakic eyes and pseudophakic eyes were not investigated so far.Further elucidation is warranted in the near future.

The immediate IOP increase after intravitreal injection may adversely affect the visual function[7,23].Lamet al[18]reported that history of glaucoma surgery reduced immediate IOP increase after intravitreal anti-VEGF agent injection and recovery time of IOP to baseline was short.Furthermore,Shoeibiet al[11]reported that administration of timolol,brimonidine, oral acetazolamide, and intravenous mannitol 30-60min before intravitreal bevacizumab injection did not reduce the immediate IOP increase.VR reduces immediate IOP increase[9-11].Chenet al[24]reported that VR might be a potential cause of endophthalmitis after intravitreal injection and its visual outcome is poor[25].VR rates should be reduced as much as possible.Anterior chamber paracentesis can reduce immediate IOP increase and is effective for retinal nerve fiber loss in anti-VEGF therapy[26].Although the onset frequency is extremely low, anterior chamber paracentesis can cause bacterial endophthalmitis and lens injuries[27].The accuracy of the intravitreal injection volume is a highly important factor for considering VR.Meyeret al[28]determined a significant variance in the accuracy, precision and repeatability of the proposed dose for intravitreal injection.Guestet al[29]reported that the IVAI volume and variability using the new prefilled syringe were significantly higher than the volume injected using the conventional BD Luer-Lok 1-mL syringe.Furthermore, Guestet al[29]concluded that the design of the prefilled syringe in which aflibercept is packaged should also be reconsidered.Both new treatment methods that reduce immediate IOP increase after IVAI without VR and accurate IVAI volume measurement are warranted in current clinical practice.

ACKNOWLEDGEMENTS

Conflicts of Interest: Muto T,None;Machida S,None;Imaizumi S,None.