Incidence and Risk Factors for Macular Hemorrhage Following Intravitreal Ranibizumab Injection for Neovascular Age-Related Macular Degeneration

Abstract

Purpose:

To analyze the incidence of and risk factors for newly developed or increased macular hemorrhage after intravitreal ranibizumab injection (IVR) for neovascular age-related macular degeneration (AMD).

Methods:

We performed a retrospective chart review of 220 subjects (220 eyes) from 5 hospitals who received IVRs for neovascular AMD between 1 June 2009 and 30 June 2010. Systemic conditions (age, sex, presence of diabetes, hypertension, cardiovascular disease, smoking, and use of anticoagulation agent) and presence of hemorrhage at the initial exam were evaluated. The primary study outcome was the incidence of newly developed or increased macular hemorrhage during the 1-month postinjection period.

Results:

The incidence of newly developed or increased macular hemorrhage including vitreous hemorrhage was 8% (18/220). Presence of diabetes was found to be a risk factor for macular hemorrhage [odds ratios (OR): 2.16, 95% confidence intervals (CI): 1.07–8.13]. When subjects had both diabetes and hypertension, the risk of macular hemorrhage after injection increased 4.8-fold (OR: 4.84, CI: 1.24–18.85).

Conclusion:

The systemic condition of subjects was found to be an important risk factor for newly developed or increased macular hemorrhage after IVR for neovascular AMD. More consideration should be given to the status of diabetes and hypertension in subjects who receive ranibizumab for neovascular AMD.

Introduction

Intravitreal ranibizumab injection (IVR) has been used as a standard treatment for neovascular age-related macular degeneration (AMD) to achieve anatomic stability and preservation or improvement of visual acuity.^1–4 However, vision loss after ranibizumab therapy occurs in a substantial portion of subjects who receive ranibizumab for neovascular AMD and has been reportedly associated with lesion characteristics such as pigmentary abnormalities, atrophic scars, and the absence of leakage.^5–7 Preservation of photoreceptor and retinal pigment epithelium function in subjects receiving ranibizumab therapy to preserve visual acuity is thus as important as the treatment itself. Newly developed or increased macular hemorrhage after IVR is one of the major ocular complications that can cause significant visual acuity loss and permanent retina damage. Until now, reports regarding the incidence of and risk factors for macular hemorrhage vary widely, and most results have been reported for eyes with different clinical characteristics. In the Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab In the Treatment of Neovascular AMD (MARINA) and Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in AMD (ANCHOR) trials, the incidence of hemorrhage is reported to be less than 1%.^1–3 However, in other studies, the incidence is reported to be up to 40%.⁸ With the exception of systemic anticoagulation, there is little information regarding the systemic and ocular risk factors for submacular hemorrhage after IVR.^9–11 The aim of this study is to report the incidence of and risk factors for macular hemorrhage after IVR for neovascular AMD.

Methods

This was a retrospective, interventional, case control, multicenter study. Patients who had received 0.5 mg of intravitreal ranibizumab (Lucentis^®; Novartis) injections for neovascular AMD between 1 June 2009 and 30 June 2010 were included. Neovascular AMD was defined as active leakage from the choroidal neovascularization (CNV) in fluoresein angiography. Ethics approval for the retrospective chart review protocol was obtained from the 5 hospitals' institutional review boards, and the study and data accumulation were carried out in adherence to the tenets of the Declaration of Helsinki.

The primary study outcome was the incidence of newly developed or increased macular hemorrhage during the 1-month postinjection period. The secondary study outcome included the risk factors associated with newly developed or increased macular hemorrhage after the injection.

Ophthalmologic evaluations

Subjects underwent a complete ocular examination at baseline and then at the 1-, 3-, and 6-month follow-up examinations. All of the eyes were followed for at least 6 months. The follow-up examinations included visual acuity measurement, intraocular pressure measurements, slit-lamp biomicroscopy, fundus examination, and color fundus photography. BCVA measurements were performed with the Snellen testing chart or ETDRS chart, and the results were converted to the logarithm of the minimum angle of resolution (LogMAR). Intraocular pressure was measured using a noncontact tonometer (Tonometer TF-X; Canon). The color fundus photographs at base line and at the 1-month follow-up visit were reviewed to assess for newly developed or increased macular hemorrhage after IVR. We defined newly developed or increased macular hemorrhage after IVR as follows: (1) a new macular hemorrhage with a greatest diameter exceeding 200 μm within 1 disc diameter of the foveal center within 1 month of injection, (2) an increase in size of a greatest diameter of preexisting macular hemorrhage over 200 μm within the temporal vascular arcades within 1 month of injection. Data interpretation was performed by 2 retinal specialists (S.S.J., S.W.M.) who were masked to the subject characteristics. When there were disagreements between the 2 graders, they discussed the grading and came to a consensus on the final diagnosis. The intergrader and intragrader agreements were κ=0.85 and 0.90, respectively, indicating good reliability.

Injection technique

All injections were administered in the usual sterile fashion with a sterile lid speculum and 5% topical povidone-iodine.¹² Topical antibiotic eye drops were prescribed for 5 to 7 days after the injections.

Systemic risk factor assessments

Possible risk factors that might influence the development of or increase in macular hemorrhage, including the subject age, sex, presence of diabetes and hypertension, smoking status, history of cardio or cerebrovascular diseases, and the use of anticoagulant/antiplatelet agents, were evaluated. Total injection number and presence of hemorrhage at initial presentation were also included for analysis. Diabetes was defined as a fasting serum glucose level ≥7 mmol/L, random glucose level ≥11 mmol/L, or current treatment for diabetes. Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or current use of antihypertensive medication. The history of cerebral and cardiovascular diseases and use of anticoagulant/antiplatelet agents were acquired from the subjects at the baseline visit.

Statistical analysis

We estimated that this study has at least 80% statistical power to detect a 25% difference in potential risk factors between the 220 subjects with or without an increase in hemorrhage. All potential risk factors were analyzed either as binary (e.g., diabetes, hypertension) or as linear traits for continuous variables (e.g., age).

We used ANOVA to assess continuous variables and the Chi-square test for binary variables to compare means and percentages. For evaluating associations with hemorrhage, ordinal logistic regression models were used to estimate Wald scores, odds ratios (OR), and 95% confidence intervals (CI) after adjusting for covariates. In multivariate-adjusted analyses, 2 models were created: model 1 was adjusted for diabetes, use of anticoagulant, smoking, and presence of hemorrhage at initial presentation; model 2 was adjusted for the factors in model 1 and diabetes with or without hypertension instead of diabetes alone. P values<0.05 were considered statistically significant. All tests of significance were 2 sided, with a significance level of 0.05.

All data were analyzed using IBM SPSS Statistics 19.0 (IBM; Armonk).

Results

A total of 286 consecutive subjects (286 eyes) were identified at 5 institutions. Of these 286 subjects, 38 (13%) were excluded because of loss during follow-up or incomplete medical records regarding systemic diseases or medication. Four subjects (1%) were excluded because of suspicious combined retinal vascular disease other than AMD, including retinal vein occlusion, retinal macroaneurysm, and CNV due to myopic degeneration. Twenty-four subjects (8%) were excluded because of their histories of previous treatment for neovascular AMD with photodynamic therapy or intravitreal injection with another anti-VEGF (e.g., bevacizumab) in the past 3 months. The remaining 220 subjects (1 eye per subject only) were the subjects of this report. The baseline demographic characteristics are summarized in Table 1. The mean age of the study subjects was 71 years, and 137 subjects (62.3%) were men. At baseline, the color fundus photographs confirmed preexisting macular hemorrhage in 129 eyes (58.6%).

Table 1.

Baseline Patient Demographics

Demographic	n	%	P value
Gender (male/female)	137/83	62.27/37.73	0.103
Hemorrhage	129/73	63.86/36.14
Nonhemorrhage	8/10	44.44/55.56
Age			0.645
Hemorrhage	70.50±8.92
Nonhemorrhage	71.50±7.60
Diabetes mellitus (no/yes)	167/53	75.91/24.09	0.045
Hemorrhage	157/45	77.72/22.28
Nonhemorrhage	10/8	55.56/44.44
Hypertension (no/yes)	111/109	50.45/49.55	0.129
Hemorrhage	105/97	51.98/48.02
Nonhemorrhage	6/12	33.33/66.67
Cardiovascular disease (no/yes)	204/16	92.73/7.27	1.000
Hemorrhage	187/15	92.57/7.43
Nonhemorrhage	17/1	94.44/5.56
Anticoagulant (no/yes)	190/30	86.36/13.64	0.079
Hemorrhage	177/25	87.62/12.38
Nonhemorrhage	13/5	72.22/27.78
Smoking history (no/yes)	207/13	94.09/5.91	1.000
Hemorrhage	190/12	94.06/5.94
Nonhemorrhage	17/1	94.44/5.56
Cataract operation history (no/yes)	159/61	72.27/27.73	1.000
Hemorrhage	146/56	72.28/27.72
Nonhemorrhage	13/5	72.22/27.78
Hemorrhage at present (no/yes)	91/129	41.36/58.64	0.222
Hemorrhage	86/116	42.57/57.43
Nonhemorrhage	5/13	27.78/72.22
PCV/not PCV	16/204	7.27/92.73	1.000
Hemorrhage	15/187	7.43/92.57
Nonhemorrhage	1/17	5.56/94.44
Total injection number			0.089
Hemorrhage	2.78±1.06
Nonhemorrhage	3.22±1.00

ANOVA to assess continuous variables and the chi-square test for binary variables.

PCV, polypoidal choroidal vasculopathy.

There were 18 eyes (8%) with newly developed or increased hemorrhage after IVR. In 6 eyes (3%), the hemorrhage was newly developed and not associated with a previous hemorrhagic lesion, and in 12 eyes (5%: 12 of 220), the hemorrhage was increased or extended from a previous hemorrhagic lesion.

Table 2 shows the results of multivariate analysis for risk factors, including presence of diabetes, hypertension, cardiovascular or cerebrovascular disease, macular hemorrhage at baseline, anticoagulant/antiplatelet usage, and smoking history. The only risk factor for newly developed or increased hemorrhage was the presence of diabetes. After multivariable adjustment, the independent risk factor for hemorrhage was the presence of diabetes (adjusted for history of diabetes, hypertension, smoking history, use of anticoagulants, and presence of macular hemorrhage at baseline visit, OR: 2.16, CI: 1.070–8.130, Table 2). In addition, there was a 4.84-fold increase in the incidence of hemorrhage in those who had both diabetes and hypertension compared with those subjects without diabetes and hypertension (OR: 4.84, CI: 1.243–18.85, Table 2). Subjects with anticoagulant/antiplatelet treatment had a higher proportion of macular hemorrhage after IVR compared with subjects without systemic anticoagulant treatment (16.7% vs. 6.8%), although the difference was not statistically significant.

Table 2.

Multivariate Analysis of Risk Factors for Hemorrhage

Independent variable	Odds ratio	95% confidence interval
Diabetes mellitus/hypertension
Diabetes mellitus (no)/hypertension (yes)	3.355	0.549–20.502
Diabetes mellitus (yes)/hypertension (no)	2.193	0.578–8.324
Diabetes mellitus (yes)/hypertension (yes)	4.840	1.243–18.850
Smoking history	1.240	0.140–11.002
Hemorrhage at present	2.544	0.823–7.871
PCV type	0.648	0.074–5.694

Adjusted for history of diabetes, hypertension, smoking history, use of anticoagulants, and presence of macular hemorrhage at baseline visit.

Discussion

There is worldwide consensus that IVR is a safe and effective treatment for neovascular AMD.^1–4,13,14 However, newly developed or progressed macular hemorrhages after IVR have been a concern to both treating physicians and patients, because it can lead to underlying photoreceptor cell damage and subsequent impairment of visual recovery. Reports regarding the incidence of intraocular hemorrhage complications related to IVR vary from a very low incidence of 0.01 to a high percentage of up to 20%.^3,8,15 In this study, the incidence of macular hemorrhage was 8%. This result is higher than in previous studies in Caucasian populations but similar to a previous Japanese study.^3,8 There is the possibility that more cases of macular hemorrhages may be found in Asian populations due to a relatively high proportion of polypoidal choroidal vasculopathy among Asian neovascular AMD eyes¹⁶; however, only a small proportion of our study subjects underwent ICG, and we could not confirm or deny this speculation. Further large prospective studies are needed to confirm whether Asian neovascular AMD subjects have a relatively higher macular hemorrhage incidence after IVR than subjects of other ethnicities.

Systemic anticoagulant treatment has been a concern for neovascular AMD subjects who receive IVR.^9–11 Kiernan et al. reported that both systemic Plavix and Coumadin use were significantly associated with an increased risk for the development of intraocular hemorrhage in subjects with neovascular AMD.⁹ However, Mason et al. reported that the risk of hemorrhagic complications in systemically anticoagulated subjects receiving intravitreal injections is extremely low.¹⁰ Although statistically not significant, our study results showed that subjects with systemic anticoagulant treatment had a higher proportion of macular hemorrhage after IVR compared with those of subjects without systemic anticoagulant treatment. There is the possibility that the small number of subjects who received systemic anticoagulant treatment in our study may have influenced the statistical result (30/220, 13.6%). While the risk of discontinuing these medicines may affect a subjects' morbidity or mortality, emphasis on appropriate medication dosage and a therapeutic international normalized ratio of these drugs should be made a subject of care.

In this study, diabetes was the only risk factor for newly developed or increased macular hemorrhage after IVR for neovascular AMD. When subjects had both diabetes and hypertension, the risk of macular hemorrhage increased 4-fold compared with that of subjects who did not have diabetes or hypertension. Diabetes and hypertension are known to be related to an increased risk of hemorrhages in various vascular pathologies.^17,18 Chronic hyperglycemia and hypertensive vascular damage result in atherosclerosis of vessels and can lead to vascular endothelium disruption and increased hemorrhagic tendency. More emphasis on systemic conditions such as diabetes and hypertension should be made for subjects who receive IVR for neovascular AMD.

The strengths of our study were as follows. First, it was a multicenter study with a relatively large number of subjects. We also identified risk factors by conducting a multivariate risk factor analysis. However, our study also has limitations. Being a retrospective, nonrandomized study, there is a possibility of selection bias. However, all subjects were followed for at least 6 months after IVR and did not receive any other treatment in the 3 months before or after the IVR. Second, there is the possibility that hemorrhages might be associated with natural disease progression. However, we include the incidence of new hemorrhages, as new or increased hemorrhages occurred only 1 month after the IVR; so, we think that the incidence is relatively accurate. Third, although the definition of development or an increase of macular hemorrhage was defined by the greatest diameter of hemorrhage, we were not able to calculate the total hemorrhage area change due to technical limitation. Lastly, there was no standardization of IVR, so there is a possibility that subjects with severe massive hemorrhages who did not receive IVR were not included in this study, which may have resulted in underestimation of the incidence of macular hemorrhage.

In conclusion, we found that the incidence of newly developed or increased macular hemorrhage, including vitreous hemorrhage, was about 8% for neovascular AMD subjects who received IVR. Systemic conditions such as diabetes and hypertension appear to be important risk factors for newly developed or increased macular hemorrhage after IVR. Eyes with use of anticoagulants had a tendency to increase hemorrhages after IVR. Results of large, prospective randomized studies are needed to confirm whether Asians have a relatively high incidence of hemorrhages after IVR.

Footnotes

Acknowledgments

This study was supported by funding (R1005961) from Novartis Korea, Seoul, Korea.

Author Disclosure Statement

None of the authors have any financial/conflicting interests to disclose. The authors have no financial conflict of interest with Novartis Korea.

References

Rosenfeld

P.J.

, Brown

D.M.

, Heier

J.S.

et al. Ranibizumab for neovascular age-related macular degeneration. N. Engl. J. Med., 355:1419e31. 2006.

Brown

D.M.

, Michels

, Kaiser

P.K.

et al. Ranibizumab versus verteporfin photodynamic therapy for neovascular agerelated macular degeneration; two year results of the ANCHOR study. Ophthalmology, 116:57–65. 2009.

Regillo

C.D.

, Brown

D.M.

, Abraham

et al. Randomized, double-masked, sham controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER Study Year 1. Am. J. Ophthalmol., 145:239e48. 2008.

Holz

F.G.

, Amoaku

, Donate

et al. SUSTAIN Study Group. Safety and efficacy of a flexible dosing regimen of ranibizumab in neovascular age-related macular degeneration: the SUSTAIN study. Ophthalmology, 118:663–671. 2011.

Mariani

, Deli

, Ambresin

et al. Characteristics of eyes with secondary loss of visual acuity receiving variable dosing ranibizumab for neovascular age-related macular degeneration. Graefes Arch. Clin. Exp. Ophthalmol., 249:1635–1642. 2011.

Bloch

S.B.

, la Cour

, Sander

et al. Predictors of 1-year visual outcome in neovascular age-related macular degeneration following intravitreal ranibizumab treatment. Acta Ophthalmol., 91:42–47. 2013.

Rosenfeld

P.J.

et al. for the MARINA and ANCHOR Study Groups. Characteristics of patients losing vision after 2 years of monthly dosing in the phase III ranibizumab clinical trials. Ophthalmology, 118:523–530. 2011.

Tano

, Ohji

EXTEND-I Study Group. Long-term efficacy and safety of ranibizumab administered pro re nata in Japanese patients with neovascular age-related macular degeneration in the EXTEND-I study. Acta Ophthalmol., May 89:208–217. 2011.

Kiernan

D.F.

, Hariprasad

S.M.

, Rusu

I.M.

et al. Epidemiology of the association between anticoagulants and intraocular hemorrhage in patients with neovascular age-related macular degeneration. Retina., 30:1573–1578. 2010.

10.

Mason

J.O.

3rd , Frederick

P.A.

, Neimkin

M.G.

et al. Incidence of hemorrhagic complications after intravitreal bevacizumab (avastin) or ranibizumab (lucentis) injections on systemically anticoagulated patients. Retina., 30:1386–1389. 2010.

11.

Levine

J.P.

, Marcus

, Sorenson

J.A.

et al. Macular hemorrhage in neovascular age-related macular degeneration after stabilization with antiangiogenic therapy. Retina., 29:1074–1079. 2009.

12.

Aiello

L.P.

, Brucker

A.J.

, Chang

et al. Evolving guidelines for intravitreous injections. Retina., 24:S3–S19. 2004.

13.

Patel

R.D.

, Momi

R.S.

, Hariprasad

S.M.

Review of ranibizumab trials for neovascular age-related macular degeneration. Semin. Ophthalmol., 26:372–379. 2011.

14.

van der Reis

M.I.

, La Heij

E.C.

, De Jong-Hesse

et al. A systematic review of the adverse events of intravitreal anti-vascular endothelial growth factor injections. Retina., 31:1449–1469. 2011.

15.

Day

, Acquah

, Mruthyunjaya

, Grossman

D.S.

, Lee

P.P.

, Sloan

F.A.

Ocular complications after anti-vascular endothelial growth factor therapy in Medicare patients with age-related macular degeneration. Am. J. Ophthalmol., 152:266–272. 2011.

16.

Kawasaki

, Yasuda

, Song

S.J.

et al. The prevalence of age-related macular degeneration in Asians. A systematic review and meta-analysis. Ophthalmology, 117:921–927. 2010.

17.

O'Donnell

M.J.

, Xavier

, Liu

et al. INTERSTROKE investigators. Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet, 376:112–123. 2010.

18.

Pezzini

, Grassi

, Del Zotto

et al. Influence of acute blood pressure on short- and mid-term outcome of ischemic and hemorrhagic stroke. J. Neurol., 258:634–640. 2011.