Intravitreal Anti-Vascular Endothelial Growth Factor for Submacular Hemorrhage from Choroidal Neovascularization

Intravitreal Anti-Vascular Endothelial Growth Factor for Submacular Hemorrhage from Choroidal Neovascularization Jae Hui Kim, MD, Young Suk Chang, MD, Jong Woo Kim, MD, Chul Gu Kim, MD, Su Jin Yoo, MD, Han Ju Cho, MD Purpose: To evaluate the efficacy of intravitreal anti-vascular endothelial growth factor (VEGF) monotherapy for patients diagnosed with exudative age-related macular degeneration (AMD) accompanied by submacular hemorrhage. Design: Retrospective, observational case series. Participants: Ninety-one eyes of 91 patients who initially presented with submacular hemorrhage associated with exudative AMD from October 2009 to September 2012. Patients were followed up for at least 6 months after treatment. Methods: Best-corrected visual acuity (BCVA) was measured at diagnosis and at 1, 3, and 6 months after treatment. The duration of symptoms was estimated. The extent of hemorrhage was estimated using fundus photography, and central foveal thickness was measured using optical coherence tomography. Change in BCVA during 6 months after treatment was estimated. The correlation of BCVA at 6 months with duration of symptoms, extent of hemorrhage, and central foveal thickness was evaluated. Main Outcome Measures: The BCVA, duration of symptoms, extent of hemorrhage, and central foveal thickness. Results: The mean duration of symptoms was 27.6
39.5 days. The mean extent of hemorrhage was 7.8
5.6 disc areas, and the mean central foveal thickness was 610.1
249.6 mm. All eyes were treated with 3.2
0.8 (range, 1e5) monthly intravitreal anti-VEGF injections during the 6-month follow-up period. The logarithm of the minimum angle of resolution BCVA at diagnosis and at 1, 3, and 6 months after the initial diagnosis was 1.38
0.53 (Snellen equivalent, 20/479), 1.27
0.57, 1.05
0.58, and 0.96
0.65 (Snellen equivalent, 20/182), respectively. The BCVA at 6 months significantly improved from baseline (P < 0.001). Poor BCVA at 6 months correlated with a longer duration of symptoms, greater extent of hemorrhage, and greater central foveal thickness (P ¼ 0.008, P ¼ 0.004, and P ¼ 0.014, respectively). Conclusions: Anti-VEGF monotherapy was found to be a useful treatment option for exudative AMD accompanied by submacular hemorrhage. However, the limited efficacy in eyes with large hemorrhage may suggest the need for more aggressive treatment in these cases. Ophthalmology 2014;121:926-935 ª 2014 by the American Academy of Ophthalmology.

The prognosis of exudative age-related macular degeneration (AMD) accompanied by submacular hemorrhage is generally poor.1e3 A variety of approaches, including vitrectomy with and without subretinal tissue plasminogen activator,4e7 photodynamic therapy,8,9 and pneumatic displacement with and without tissue plasminogen activator,10e13 have been evaluated with regard to their efficacy in promoting visual recovery. However, the effects of these treatment modalities have generally been limited, and there remains no “gold standard” for treatment. Anti-vascular endothelial growth factor (VEGF) has demonstrated excellent efficacy in the treatment of exudative AMD, regardless of the subtype of AMD.14e18 However, only limited knowledge is available regarding the efficacy of anti-VEGF monotherapy in treating exudative AMD accompanied by submacular hemorrhage. Although previous studies showed improvement in visual acuity after

926

 2014 by the American Academy of Ophthalmology Published by Elsevier Inc.

anti-VEGF monotherapy in eyes with submacular hemorrhage, the conclusions were drawn on the basis of relatively small study populations.19e23 Moreover, the differences in outcomes among subtypes of exudative AMD have not yet been addressed. In this study, we report the 6-month outcome of intravitreal anti-VEGF therapy for 91 patients who were diagnosed with exudative AMD accompanied by submacular hemorrhage. We also analyzed the prognostic factors predictive of visual acuity outcomes and difference in visual outcome among subtypes of exudative AMD.

Methods This retrospective, observational case series was performed at a single center according to the tenets of the Declaration of Helsinki. The study was approved by the institutional review board (No. ISSN 0161-6420/14/$ - see front matter http://dx.doi.org/10.1016/j.ophtha.2013.11.004

Kim et al



Treatment of Submacular Hemorrhage

A-2013-011). We conducted a computerized search for patients who were newly diagnosed with exudative AMD from September 2009 to November 2012 at Kim’s Eye Hospital. Some of the patients who were included in the previous study23 that reported treatment outcome of submacular hemorrhage secondary to polypoidal choroidal vasculopathy (PCV) were also included in the current observational study. However, we did not duplicate the data of the previous study, and all of the evaluations were newly performed by independent examiners. Patients aged older than 50 years were included. In addition, patients with foveainvolving subretinal hemorrhage extending over at least 50% of the lesion or at least 3 disc areas were included. Patients who exhibited initial visual acuity 20/30 and were treated with a monthly injection of intravitreal ranibizumab were included. In general, treatment for submacular hemorrhage was administered except when the patient strongly denied it. The possibility of systemic adverse events was sufficiently discussed with patients who had a recent history of cardiovascular/cerebrovascular events or patients with a relatively poor general condition before the treatment. Eyes previously diagnosed with exudative AMD and under treatment at the time of the study were excluded. That is, only newly diagnosed, treatment-naïve eyes were included. One of the authors (J.H.K.) conducted a search for eligible patients by reviewing the fundus photographs and medical records of every patient. To be included in this study, all subjects were required to undergo a comprehensive ophthalmologic examination, including measurements of best-corrected visual acuity (BCVA), 90-diopter lens slit-lamp biomicroscopy, fundus photography, fluorescein angiography, and spectral domain optical coherence tomography (OCT) (Spectral OCT/SLO; OTI Ophthalmic Technologies Inc., Miami, FL). Indocyanine green (ICG) angiography with a confocal laser-scanning system (HRA-2; Heidelberg Engineering, Dossenheim, Germany) was performed at the discretion of each physician. The exclusion criteria included less than 6 months of follow-up; duration of symptoms longer than 6 months; severe media opacity; evidence of end-stage AMD, such as central geographic atrophy or disciform scarring; evidence of a macroaneurysm; proliferative diabetic retinopathy; central retinal vascular occlusion; or history of intraocular surgery except cataract surgery. Eyes that had undergone pneumatic displacement or photodynamic therapy and eyes that had received intravitreal injections of tissue plasminogen activator were excluded. The number of eyes that underwent vitrectomy during the follow-up period because of the development of severe vitreous hemorrhage was recorded. However, any data pertaining to these eyes were excluded from the result analysis. If submacular hemorrhage developed in both eyes, the eye that had been affected first was included. That is, only 1 eye was included in 1 patient. The main outcome measures were the BCVA, duration of symptoms, extent of hemorrhage, and central foveal thickness, which may reflect the depth of a subretinal hemorrhage. Visual acuities were converted to logarithm of the minimum angle of resolution values for analysis. The extent of hemorrhage was estimated using disc area as the basic unit. Central foveal thickness was defined as the distance between the internal limiting membrane and the Bruch’s membrane at the fovea and was manually measured using the calipers provided by an OCT software program. In this study, we considered the possibility that measurements of hemorrhages exceeding 20 disc areas and central foveal thickness exceeding 1500 mm could be inaccurate. Thus, 20 disc areas of hemorrhage and 1500 mm of central foveal thickness were considered threshold values; lesions exceeding this range were considered 20 disc areas in size or 1500 mm in depth. The extent of hemorrhage and central foveal thickness were estimated by a single examiner (J.H.K.).

The results of ICG angiography were analyzed by 2 independent examiners (J.H.K. and Y.S.C.). Cases of exudative AMD were classified as typical exudative AMD or PCV on the basis of the ICG findings. Cases exhibiting branching vascular networks or terminating polypoidal lesions were diagnosed with PCV. The other cases were classified as typical exudative AMD. Any disagreement was settled by discussion between the 2 examiners. The analyses that were performed are described in the following sections.

Changes in Best-Corrected Visual Acuity and Factors Associated with Best-Corrected Visual Acuity at 6 Months Baseline BCVA was compared with BCVA at 1, 3, and 6 months. Additional analysis was performed for cases that presented for the 12-month follow-up examination. Baseline BCVA, 6-month BCVA, and 12-month BCVA were compared. Changes in the BCVA during the 6 months also were analyzed using a mixed model; data from eyes lost to follow-up were not excluded from this additional analysis. The association of BCVA at 6 months with baseline BCVA, duration of symptoms, extent of hemorrhage, and central foveal thickness was analyzed.

Changes in Extent of Hemorrhage and Central Foveal Thickness Baseline extent of hemorrhage and central foveal thickness were compared with those values at 3 and 6 months. Additional analysis was performed for cases that presented for the 12-month follow-up examination. In these cases, baseline values, BCVA values at 3 and 6 months, and values at 12 months also were compared. The cutoff value of the baseline extent of hemorrhage and central foveal thickness for predicting a 6-month visual acuity of 20/200 was estimated.

Comparison of Eyes with Typical Exudative Agerelated Macular Degeneration, Eyes with Polypoidal Choroidal Vasculopathy, and Unclassified Eyes The baseline and 6-month values for BCVA, extent of hemorrhage, and central foveal thickness were compared between the eyes diagnosed with typical exudative AMD, eyes diagnosed with PCV, and unclassified eyes. The 6-month changes in BCVA, extent of hemorrhage, and central foveal thickness also were estimated within each group.

Comparison with Eyes that Developed Submacular Hemorrhage during Follow-up Eyes were classified into the hemorrhage during follow-up group if submacular hemorrhage was not observed at the diagnosis of exudative AMD but developed during the follow-up period. For comparison, 91 patients who were originally included in the present study were classified into the hemorrhage at diagnosis group. The BCVA before the development of hemorrhage, BCVA at the development of hemorrhage, and BCVA at 1, 3, and 6 months after the development of hemorrhage were compared. In addition, the change in BCVA was compared between the 2 groups (hemorrhage during follow-up group vs. hemorrhage at diagnosis group) as follows: In the hemorrhage during follow-up group, changes between the BCVA that were measured before the development of hemorrhage and the BCVA that was measured 6 months after the development of hemorrhage were defined as the changes in the BCVA. In the hemorrhage at diagnosis group, the changes between

927

Ophthalmology Volume 121, Number 4, April 2014 the BCVA that was measured when the hemorrhage developed and the BCVA that was measured at 6 months after the development of hemorrhage were defined as the changes in BCVA.

Other Analyses The proportion of eyes in each diagnosis (typical exudative AMD vs. PCV vs. unclassified) was compared between the included eyes and the excluded eyes. The extent of hemorrhage, central foveal thickness, BCVA at diagnosis, and BCVA at 6 months were compared between patients in whom anticoagulants were or were not used. The association between the number of anti-VEGF injections and the BCVA at 6 months or the changes in the BCVA during the 6 months was analyzed. The baseline extent of hemorrhage, the baseline central foveal thickness, and the number of anti-VEGF injections during the first 6 months were estimated in eyes requiring vitrectomy because of vitreous hemorrhage during the 6-month follow-up period. Statistical analyses were performed with a commercially available software package (SPSS v. 12.0 for Windows; SPSS Inc., Chicago, IL). Differences among various time points were analyzed by using a repeated-measures analysis of variance, and individual comparisons were performed using Bonferroni’s method. Additional analysis for BCVA changes during the 6-month follow-up period was performed using a mixed model. Differences between the defined groups were analyzed using 1-way analyses of variance with the Tukey test or independent samples t test. Pearson correlation analysis was performed to verify any associations between variables. Comparison of the proportion of eyes in each diagnosis between included and excluded eyes was performed using the chi-square test. The cutoff values were estimated using receiver operating characteristic curve analysis. A P value <0.05 was considered significant.

Results A total of 1573 patients were newly diagnosed with exudative AMD during the aforementioned period. Among them, 159 patients (10.1%) initially presented with submacular hemorrhage. Among the 159 affected eyes, 68 (42.8%) were excluded for the following reasons: other treatment (29 eyes), less than 6 months of follow-up (23 eyes), vitrectomy during the follow-up period (11 eyes), untreated (2 eyes), or other reasons (3 eyes). Ultimately, 91 eyes of 91 patients (57.2%) were included in the statistical analysis (Table 1). Fifty-nine patients (64.8%) were male, and 32 patients (35.2%) were female. The mean (
standard deviation) age was 69.9
8.4 years (range, 51e86 years), and the mean spherical equivalent was þ0.35
1.28 diopters (3.75 to þ3.00 diopters). The mean duration of symptoms was 27.6
39.5 days (range, 1e180 days). The mean BCVA was 1.38
0.53 (Snellen equivalent, 20/479; range, hand motion to 20/30). The mean extent of hemorrhage was 7.8
5.6 disc areas (range, 3e20 disc areas), and the mean central foveal thickness was 610.1
249.6 mm (range, 261e1500 mm). The BCVA at diagnosis, duration of symptoms, extent of hemorrhage, and central foveal thickness in 68 excluded eyes were 1.50
0.56 (Snellen equivalent, 20/632; range, hand motion to 20/50), 23.8
37.1 days, 10.4
6.1 disc areas, and 671.9
271.8 mm, respectively. When compared with the 91 eyes included in the study, the excluded eyes exhibited a greater extent of hemorrhage and greater central foveal thickness (P ¼ 0.005 and P ¼ 0.005, respectively). Although the BCVA at diagnosis was slightly worse and the duration of symptoms was slightly shorter in the excluded eyes, the difference was not significant (P ¼ 0.153 and P ¼ 0.545, respectively). Indocyanine green angiography results at the time of diagnosis were available in 88 eyes (96.7%). Among them, 23 eyes (26.1%)

928

Table 1. Baseline Characteristics of Patients with Exudative AgeRelated Macular Degeneration with Large Submacular Hemorrhage as an Initial Presentation (N ¼ 91) Characteristic Age (yrs) Sex, no. (%) Male Female Diagnosis, no. (%)* Typical exudative AMD PCV Definitive diagnosis was not possible BCVA (logMAR) Extent of hemorrhage (disc areas) Central foveal thickness (mm)

69.9
8.4 (range, 51e86) 59 (64.8%) 32 (35.2%) 23 (26.1%) 42 (47.7%) 23 (26.1%) 1.38
0.53 (range, HM to 20/30) 7.8
5.6 (range, 3e20) 610.1
249.6 (range, 261e1500)

AMD ¼ age-related macular degeneration; BCVA ¼ best-corrected visual acuity; HM ¼ hand motion; logMAR ¼ logarithm of the minimum angle of resolution; PCV ¼ polypoidal choroidal vasculopathy; SD ¼ standard deviation. Data are mean
SD (range) unless otherwise indicated. *Data were derived from 88 eyes for which indocyanine green angiography results from the time of diagnosis were available.

and 42 eyes (47.7%) were ultimately diagnosed with typical exudative AMD and PCV, respectively. A definite diagnosis was not possible in the remaining 23 cases (26.1%) because thick subretinal hemorrhage precluded reliable ICG angiography images of the lesion. The mean extent of hemorrhage for eyes with and without definitive diagnoses was 7.0
5.0 versus 10.5
6.9 disc areas, the central foveal thickness was 580.8
225.1 versus 716.1
293.5 mm, and the BCVA at 6 months was 0.94
0.67 and 1.11
0.60, respectively. The ICG results were available for 60 of 68 of the excluded eyes (88.2%). Eleven eyes (18.3%) and 32 eyes (53.3%) were ultimately diagnosed with typical exudative AMD and PCV, respectively. A definite diagnosis was not possible in the remaining 17 eyes (28.3%). No significant difference among each diagnosis was found between the included eyes (N ¼ 88) and the excluded eyes (N ¼ 60; P ¼ 0.540).

Changes in Best-Corrected Visual Acuity The patients were treated with 3.2
0.8 (range, 1e5) monthly intravitreal anti-VEGF injections during the 6-month follow-up period. Three eyes were treated with bevacizumab only. The other 3 eyes were treated with both ranibizumab and bevacizumab. The remaining 85 eyes were treated with ranibizumab only during the 6-month follow-up period. Figure 1 shows the resolution of submacular hemorrhage as observed at each time point. The BCVA at baseline and 1, 3, and 6 months after the initial diagnosis was 1.38
0.53 (Snellen equivalent, 20/479), 1.27
0.57 (Snellen equivalent, 20/372), 1.05
0.58 (Snellen equivalent, 20/224), and 0.96
0.65 (Snellen equivalent, 20/182; range, hand motion to 20/20), respectively (Fig 2A). The BCVAs differed significantly among the 4 time points (P < 0.001). The BCVA at diagnosis was significantly different from that measured at 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively), whereas the difference was not significantly different from that measured at 1 month after the diagnosis (P ¼ 0.064). An improvement of 3 lines or more in BCVA was noted in 54 eyes (59.3%). A deterioration of

Kim et al



Treatment of Submacular Hemorrhage

Figure 1. Fundus photography (A, C) and optical coherence tomography (B, D) findings of an eye with extensive submacular hemorrhage secondary to exudative age-related macular degeneration. At the time of diagnosis, visual acuity was measured as 20/400 (A, B). After 3 monthly ranibizumab injections during the 6-month follow-up period, the submacular hemorrhage had resolved completely and visual acuity had improved to 20/30 (C, D).

3 lines or more was noted in 8 eyes (8.8%). The remaining 29 eyes (31.9%) exhibited stable BCVA (Table 2). The significant improvement in BCVA during the 6-month period was noted in an additional analysis with 114 patients, 23 of whom were lost to follow-up and 91 of whom completed 6 months of follow-up (P < 0.001).

Factors Associated with Best-Corrected Visual Acuity at 6 Months Baseline BCVA was significantly associated with BCVA at the 6month follow-up (P < 0.001; r ¼ 0.494). The duration of symptoms was significantly associated with BCVA at 6 months (P ¼ 0.008; r ¼ 0.278), whereas the association with baseline BCVA was not significant. The extent of hemorrhage was significantly associated with BCVA at 6 months (P ¼ 0.004; r ¼ 0.301), whereas the association with visual acuity at diagnosis was not significant (P ¼ 0.084). The baseline central foveal thickness was significantly associated with both the baseline BCVA and the BCVA at 6 months (P ¼ 0.031; r ¼ 0.223 and P ¼ 0.014; r ¼ 0.253, respectively). The association between the number of antiVEGF injections was not significantly associated with the BCVA at 6 months (P ¼ 0.131) or changes in the BCVA during the 6 months (P ¼ 0.321).

diagnosis was significantly different from that measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). The mean central foveal thicknesses at baseline and 3 and 6 months after the initial diagnosis were 596.8
226.1, 312.1
168.2, and 282.9
139.3 mm, respectively (Fig 3B), and the central foveal thickness differed significantly among the 3 time points (P < 0.001). The value at diagnosis was significantly different from that measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). The BCVA at 6 months was significantly associated with the central foveal thickness at 6 months (P ¼ 0.033), whereas the association with the extent of hemorrhage at 6 months was not significant (P ¼ 0.202). The changes in BCVA during the 6 months after treatment were not significantly associated with the changes in the extent of hemorrhage (P ¼ 0.091) or the central foveal thickness (P ¼ 0.081). The threshold value of the extent of hemorrhage was 9 disc areas (sensitivity ¼ 0.73, specificity ¼ 0.43), and the threshold value of the central foveal thickness was 704 mm (sensitivity ¼ 0.77, specificity ¼ 0.41). Thus, 20/200 or worse visual acuity at 6 months may be predicted in eyes exhibiting an extent of hemorrhage of 9 disc areas or more or a central foveal thickness of 704 mm or more at diagnosis.

Changes in the Extent of Hemorrhage and Central Foveal Thickness

Comparison of Eyes with Typical Exudative Agerelated Macular Degeneration, Eyes with Polypoidal Choroidal Vasculopathy, and Unclassified Eyes

Both fundus photography and OCT data 3 and 6 months after the initial diagnosis were available in 73 of 91 patients (80.2%). The association between the extent of hemorrhage and the central foveal thickness at diagnosis was not significant (P ¼ 0.546). Complete resolution of hemorrhage within 3 and 6 months was noted in 33 (45.2%) and 49 (67.1%) of these patients, respectively. The mean extent of hemorrhage at baseline and at 3 and 6 months after the initial diagnosis was 7.5
5.5, 2.3
3.2, and 1.1
1.9 disc areas, respectively (Fig 3A). The extent of hemorrhage differed significantly among the 3 time points (P < 0.001). The value at

The features at baseline and 6 months among the diagnosis groups are summarized in Table 3. In the typical exudative AMD group (N ¼ 23), the extent of hemorrhage and the central foveal thickness at diagnosis were 6.4
5.1 disc areas and 511.7
182.4 mm, respectively. In the PCV group (N ¼ 42), these values were 7.2
4.9 disc areas and 618.6
238.9 mm, respectively. In the unclassified group (N ¼ 23), these values were 10.5
6.8 disc areas and 716.1
293.5 mm, respectively. In the typical exudative AMD group, the BCVAs at baseline and 1, 3, and 6 months after the initial diagnosis were 1.33
0.55

929

Ophthalmology Volume 121, Number 4, April 2014 Table 2. Distribution of Eyes According to the Change in Visual Acuity Degree of Change

Month 6*

Month 12y

Gain 3 lines Stable Loss 3 lines

54 (59.3%) 29 (31.9%) 8 (8.8%)

46 (64.8%) 21 (29.6%) 4 (5.6%)

*Data were obtained from 91 eyes that finished the 6-month follow-up program. y Data were obtained from 71 eyes that finished the 12-month follow-up program.

(P ¼ 0.001 and P < 0.001, respectively), whereas the difference was not significantly different from that measured 1 month after the diagnosis (P ¼ 0.374). In the unclassified group, the BCVAs at baseline and 1, 3, and 6 months after the initial diagnosis were 1.43
0.53 (Snellen equivalent, 20/538), 1.41
0.53 (Snellen equivalent, 20/514), 1.30
0.58 (Snellen equivalent, 20/339), and

Figure 2. A, Changes in the mean logarithm of the minimum angle of resolution best-corrected visual acuity of eyes that received anti-vascular endothelial growth factor monotherapy for submacular hemorrhage secondary to exudative age-related macular degeneration (AMD), according to the follow-up period. A, Solid line (closed circle) indicates eyes that completed the 6-month follow-up (N ¼ 91); dashed line (closed square) indicates eyes that completed the 12-month follow-up (N ¼ 71). B, Changes in values when the patients were divided into 3 groups according to the diagnosis. Solid line (closed circle) indicates eyes diagnosed with typical exudative AMD (N ¼ 23); dashed line (closed square) indicates eyes diagnosed with polypoidal choroidal vasculopathy (N ¼ 42); dashed-dotted line (closed triangle) indicates unclassified eyes (N ¼ 23). Statistical analysis was performed using repeated-measures analysis of variances with Bonferroni’s correction. logMAR ¼ logarithm of the minimum angle of resolution; M ¼ month.

(Snellen equivalents, 20/427), 1.21
0.61 (Snellen equivalent, 20/ 324), 0.95
0.55 (Snellen equivalent, 20/178), and 0.89
0.65 (Snellen equivalent, 20/155), respectively (Fig 2B). The BCVAs differed significantly among the 4 time points (P ¼ 0.026), and the BCVA at diagnosis was significantly different from that measured 3 and 6 months after the diagnosis (P ¼ 0.015 and P ¼ 0.012, respectively). In the PCV group, the BCVAs at baseline and 1, 3, and 6 months after the initial diagnosis were 1.41
0.53 (Snellen equivalent, 20/514), 1.27
0.58 (Snellen equivalent, 20/372), 1.00
0.58 (Snellen equivalent, 20/200), and 0.92
0.66 (Snellen equivalent, 20/166), respectively (Fig 2B). The BCVAs differed significantly among the 4 time points (P < 0.001). The BCVA at diagnosis was significantly different from that measured 3 and 6 months after the diagnosis

930

Figure 3. Changes in the extent of hemorrhage (A) and central foveal thickness (B) of eyes that received anti-vascular endothelial growth factor monotherapy for submacular hemorrhage secondary to exudative agerelated macular degeneration, according to the follow-up period. Solid lines (closed circles) indicate eyes that completed the 6-month follow-up (N ¼ 73); dashed lines (closed squares) indicate eyes that completed the 12-month follow-up (N ¼ 46). Statistical analysis was performed using repeated-measures analysis of variances with Bonferroni’s correction. M ¼ month.

Kim et al



Treatment of Submacular Hemorrhage

Table 3. Characteristics of 88 Patients with Indocyanine Green Angiography Results Available from the Time of Diagnosis and Divided into 3 Groups According to the Diagnosis: Typical Exudative Age-Related Macular Degeneration (AMD), Polypoidal Choroidal Vasculopathy (PCV), and Unclassified Features Baseline BCVA (logMAR) Extent of hemorrhage (disc areas)* Central foveal thickness (mm)* At 6 mos follow-up BCVA (logMAR) Extent of hemorrhage (disc areas)* Central foveal thickness (mm)*

P Valuey

Typical Exudative AMD (N [ 23)

PCV (N [ 42)

Unclassified (N [ 23)

1.33
0.55 6.4
5.1 511.7
182.4

1.41
0.53 7.2
4.9 618.6
238.9

1.43
0.53 10.5
6.8 716.1
293.5

0.833 0.025 0.014

0.89
0.65 0.3
30.8 281.5
118.1

0.92
0.66 1.2
2.1 268.3
136.8

1.18
0.62 2.1
2.4 329.5
156.5

0.230 0.036 0.360

BCVA ¼ best-corrected visual acuity; logMAR ¼ logarithm of the minimum angle of resolution; SD ¼ standard deviation. Data are mean
standard deviation. *Data were derived from 73 eyes for which both fundus photography and optical coherence tomography results were available at both 3 and 6 months. y Statistical analysis was performed using 1-way analysis of variance.

1.18
0.62 (Snellen equivalent, 20/302), respectively (Fig 2B). The BCVAs differed significantly among the 4 time points (P ¼ 0.021). The BCVA at diagnosis was significantly different from that measured 6 months after the diagnosis (P ¼ 0.031), whereas the difference was not significantly different from that measured 1 and 3 months after the diagnosis (P ¼ 0.229 and P ¼ 0.102, respectively). Although the baseline BCVA was slightly worse in the unclassified eyes, the difference was not significant among the 3 groups (P ¼ 0.833); however, significant differences were observed in the baseline extent of hemorrhage (P ¼ 0.025) and central foveal thickness (P ¼ 0.018). The unclassified group showed a significantly greater extent of hemorrhage (P ¼ 0.034) and greater central foveal thickness (P ¼ 0.014) than the typical exudative AMD group. Further, the PCV group showed a significantly greater central foveal thickness than the typical exudative AMD group (P ¼ 0.044). No significant difference in the BCVA at 6 months was observed among the 3 groups (P ¼ 0.223). In the typical exudative AMD group, both fundus photography and OCT data 3 and 6 months after the initial diagnosis were available for 19 of 23 patients (82.6%). In these 19 eyes, the mean extent of hemorrhage at baseline and 3 and 6 months after the initial diagnosis was 5.4
4.6, 1.1
2.2, and 0.3
0.8 disc areas, respectively. The extent of hemorrhage differed significantly among the 3 time points (P < 0.001), and the value at diagnosis was significantly different from that measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). The mean central foveal thicknesses at baseline and 3 and 6 months after the initial diagnosis were 499.3
156.5, 304.1
143.2, and 281.5
118.1 mm, respectively. The central foveal thickness differed significantly among the 3 time points (P < 0.001), and the value at diagnosis was significantly different from that measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). Both fundus photography and OCT data 3 and 6 months after the initial diagnosis were available in 35 of 42 patients (83.3%) in the PCV group. The mean extent of hemorrhage baseline and 3 and 6 months after the initial diagnosis was 7.4
5.2, 1.9
2.9, and 1.2
2.1 disc areas, respectively, in these 34 eyes. Significant differences were observed in the extent of hemorrhage among the 3 time points (P < 0.001) and between the value at diagnosis and those measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). The mean central foveal thicknesses at baseline and 3 and 6 months after the initial diagnosis were 578.2
194.1, 292.6
151.8, and 268.3
136.8 mm, respectively. The central foveal thickness differed significantly

among the 3 time points (P < 0.001), and the value at diagnosis was significantly different from that measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). Both fundus photography and OCT data 3 and 6 months after the initial diagnosis were available in 19 of 23 patients (82.6%) in the unclassified group, and the mean extent of hemorrhage at baseline and 3 and 6 months after the initial diagnosis was 9.7
6.5, 3.9
3.8, and 1.9
2.4 disc areas, respectively, in these 19 eyes. The extent of hemorrhage differed significantly among the 3 time points (P < 0.001), and the value at diagnosis was significantly different from that measured at 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). The mean central foveal thicknesses baseline and 3 and 6 months after the initial diagnosis were 732.2
283.3, 369.3
211.7, and 325.1
159.5 mm, respectively. The central foveal thickness differed significantly among the 3 time points (P < 0.001), and the value at diagnosis was significantly different from that measured 3 or 6 months after the diagnosis (P < 0.001 and P < 0.001, respectively). The anatomic outcome was compared in 73 eyes with available fundus photography and OCT results. Although the difference in BCVA (P ¼ 0.230) and central foveal thickness (P ¼ 0.360) was not significant among the 3 groups at 6 months, a significant difference in the extent of hemorrhage was observed at this time point (P ¼ 0.036). The unclassified group showed a significantly greater extent of hemorrhage at 6 months than the typical exudative AMD group (P ¼ 0.028).

Results Obtained at 12-Month Follow-up Data from the 12-month follow-up examination were available for 73 eyes. However, 2 of them underwent vitrectomy at some point during the 6- to 12-month follow-up period. Thus, the 12-month data included the data from 71 eyes (78.0%). The patients received a mean of 3.6
1.2 injections (range, 1e7) of anti-VEGF during the 12month follow-up period. In these eyes, the BCVAs at diagnosis and at 6- and 12-month follow-ups were 1.42
0.52 (Snellen equivalent, 20/526), 0.93
0.64 (Snellen equivalent, 20/170), and 0.94
0.66 (Snellen equivalent, 20/174), respectively (Fig 2A). The BCVA measured at the 12-month follow-up was significantly better than that measured at the time of diagnosis (P < 0.001), whereas the difference was not significant when compared with BCVA at 6 months (P ¼ 1.000). Three lines or greater improvement in BCVA was noted in 46 eyes (64.8%), whereas deterioration of more than 3 lines was noted in 4 eyes (5.6%). The remaining 21 eyes (29.6%) exhibited stable BCVA (Table 2).

931

Ophthalmology Volume 121, Number 4, April 2014 Both fundus photography and OCT data 3, 6, and 12 months after the initial diagnosis were available in 46 of 71 patients (64.8%), and complete resolution of hemorrhage was observed in 33 of these 46 patients (71.7%). Enlargement of the extent of hemorrhage was noted in 9 eyes (19.6%). The mean extent of hemorrhage at baseline and 3, 6, and 12 months after the initial diagnosis was 7.6
5.4, 2.4
3.3, 1.4
2.2, and 1.1
2.5 disc areas, respectively (Fig 3A). The extent of hemorrhage differed significantly among the 4 time points (P < 0.001), and the value at diagnosis was significantly different from that measured 3, 6, and 12 months after the diagnosis (P < 0.001 for all 3 time points). The mean central foveal thicknesses at baseline and 3, 6, and 12 months after the initial diagnosis were 580.4
253.3, 345.9
182.6, 305.9
148.5, and 324.8
142.1 mm, respectively (Fig 3B); these values were significantly different among the 4 time points (P < 0.001). The value at diagnosis was significantly different from that measured 3, 6, and 12 months after the diagnosis (P < 0.001 for all 3 time points).

Comparison Based on Anticoagulant Use At the time of diagnosis, 23 patients (25.3%) were using an anticoagulant. The mean extent of hemorrhage and central foveal thickness were 9.3
5.3 disc areas and 615.5
266.2 mm, respectively, in patients who were using anticoagulants and 7.2
5.7 disc areas and 608.3
245.7 mm, respectively, in patients who were not using anticoagulants. The extent of hemorrhage (P ¼ 0.137) and the central foveal thickness (P ¼ 0.906) were not different between patients who were or were not using an anticoagulant. The mean BCVAs at diagnosis and at 6 months were 1.46
0.53 and 1.12
0.66, respectively, in patients who were using anticoagulants and 1.35
0.53 and 0.91
0.64, respectively, in patients who were not using anticoagulants. The BCVAs at diagnosis (P ¼ 0.357) and at 6 months (P ¼ 0.180) were not significantly different between the 2 groups.

Comparison with Eyes that Developed Submacular Hemorrhage during Follow-up Among the 1414 patients in whom submacular hemorrhage was not observed at the initial visit, we verified the development of fovea-involving subretinal hemorrhage extending over at least 50% of the lesion or at least 3 disc areas in 82 eyes. Among these 82 eyes, 43 were excluded for various reasons, including loss to follow-up, other treatments, or evidence of end-stage AMD. Therefore, we analyzed 6-month visual outcome for 39 eyes, and these eyes were included in the hemorrhage during follow-up group. Submacular hemorrhage developed at a mean of 15.4
9.4 months (range, 1e32 months) after the initial diagnosis of exudative AMD. The mean extent of hemorrhage and central foveal thickness were 11.0
5.8 disc areas (range, 3e20 disc areas) and 568.3
256.1 mm (range, 269e1500 mm), respectively. The visual acuity before the development of hemorrhage was measured 2.9
1.5 months (range, 1e6 months) before the development of hemorrhage, and these patients were treated with 2.9
1.0 (range, 1e5) monthly intravitreal anti-VEGF injections during the 6 months after the development of hemorrhage. The BCVAs before the development of hemorrhage, at the time of the development of hemorrhage, and 3 and 6 months after the development of hemorrhage were 0.63
0.44 (Snellen equivalent, 20/85), 1.35
0.56 (Snellen equivalent, 20/447), 1.09
0.57 (Snellen equivalent, 20/246), and 1.02
0.56 (Snellen equivalent, 20/209), respectively. The BCVAs differed significantly among the 4 time points (P < 0.001). Although the BCVA significantly improved between the time that the submacular hemorrhage developed and

932

the 6 months of follow-up (P < 0.001), the 6-month BCVA showed significant deterioration when compared with the value before the development of hemorrhage (P < 0.001). When the BCVA before the development of hemorrhage was compared with the BCVA at 6 months after the development of hemorrhage, a deterioration of 3 lines or more was noted in 23 eyes (52.3%), and the BCVA remained stable (<3 lines of change) in the remaining 21 eyes (47.7%). A significant difference in the BCVA changes was observed between the hemorrhage during follow-up group (N ¼ 43) and the hemorrhage at diagnosis group (N ¼ 91) (P < 0.001).

Development of Vitreous Hemorrhage or Other Severe Ocular Adverse Events As mentioned earlier, 11 eyes required vitrectomy during the 6-month follow-up period because of the development of severe vitreous hemorrhage. In these eyes, the extent of hemorrhage, central foveal thickness, and number of anti-VEGF injections during the first 6 months were 15.3
5.4 disc areas, 684.0
354.3 mm, and 2.6
0.5 injections, respectively. No other severe ocular event, including endophthalmitis or retinal detachment, was noted.

Discussion In this study, visual acuity improved significantly after antiVEGF monotherapy in eyes with exudative AMD accompanied by submacular hemorrhage at the initial presentation. To our knowledge, this is the largest case series to evaluate the efficacy of anti-VEGF monotherapy in the treatment of exudative AMD accompanied by submacular hemorrhage. An improvement in the BCVA of 3 lines or stable vision was observed in 59% and 32% of eyes, respectively. As a result, the mean visual acuity was improved from 20/479 at baseline to 20/182 at the 6-month follow-up. No instance of a serious ocular or systemic adverse event was noted. Both the extent of hemorrhage and the central foveal thickness measured at the time of diagnosis were found to be prognostic factors of visual acuity at 6 months. No difference in visual acuity outcome between typical exudative AMD and PCV was identified. Visual acuity was significantly improved after anti-VEGF therapy in this study, whereas the mean visual acuity generally decreased in observational studies without treatment.1e3,7 In addition, the visual outcome of the present study is superior to that reported in previous studies showing the outcome of surgery7 or photodynamic therapy,19 suggesting that antiVEGF monotherapy is a useful treatment option. In the Submacular Surgery Trials (SST),7 in which the outcome of observation and surgery was compared, improvement of visual acuity by 3 lines or more during the 6 months was achieved in only 10% of eyes in the observation group and in 11% of eyes in the surgery group, whereas the proportion was 59% in our study. In addition, worsening of visual acuity by 4 lines or more was observed in 41% of eyes in the observation group and in 31% of eyes in the surgery group in the SST study, whereas only 9% of the patients in our study showed worsening of visual acuity by 3 lines or more. Although photodynamic therapy was found to effectively minimize visual loss in a certain proportion of eyes, only 27% of eyes experienced an improvement of

Kim et al



Treatment of Submacular Hemorrhage

visual acuity by 2 lines or greater and 47% of eyes experienced a worsening of visual acuity by 2 lines or greater during the 6-month period.19 The mean visual acuity at 9.4 months (mean, 20/250) was similar to that at the initial diagnosis (mean, 20/210). A relatively greater proportion of eyes exhibited resolution of subretinal hemorrhage at 6 months in the present study (67%) compared with the photodynamic therapy study (48%).19 However, the obvious differences in the baseline features between the present study and the previous studies make direct comparison of these results difficult. The baseline visual acuity of the present study (mean, 20/479) was relatively worse than that of the SST study7 (median, 20/ 200 [observation group] and 20/250 [surgery group]) and the photodynamic therapy study19 (mean, 20/210). Furthermore, the patients included in the present study were relatively younger (mean, 69.9 years) compared with the patients in the SST (median, 79 years) and the photodynamic therapy study (mean, 77 years). Finally, the extent of hemorrhage was relatively smaller in the present study (w8 disc areas); in contrast, more than 40% of eyes exhibited lesion sizes larger than 16 disc areas in the SST and the mean size of the subretinal hemorrhages was 13 disc diameters in the photodynamic therapy study. A more controlled study that directly compares the efficacy of different treatment modalities is needed to determine whether the efficacy of anti-VEGF monotherapy is superior to other modalities. When compared with previous studies using anti-VEGF monotherapy, the baseline visual acuity of our patients (mean, 20/479) was generally poorer than that reported by previous studies (mean, 20/320 to 6/34) investigating the effect of anti-VEGF monotherapy under similar conditions.19e23 However, the degree of improvement in visual acuity described in the current report was generally comparable to20,22,23 or better than19,21 that reported previously. More than 90% of included eyes exhibited improved or stable visual acuity at the 6-month follow-up. In addition, 94% of the eyes that were followed up for 12 months or more exhibited improved or stable visual acuity throughout the 12-month follow-up period. Most of our patients were treated with ranibizumab monotherapy during the first 6 months (6.6% of patients were treated with bevacizumab only or combination therapy with ranibizumab and bevacizumab). However, considering the results of recent studies reporting that there is no significant difference in efficacy between ranibizumab and bevacizumab,24,25 our results may hold true for various types of anti-VEGF treatments. A recent study showed a positive association between long symptom duration and poor visual prognosis in eyes treated with anti-VEGF therapy.22 Although the extent and thickness of the hemorrhage were found to be associated with visual outcome in previous observational studies,2,3 the predictive value of these 2 factors has not been investigated in eyes that underwent anti-VEGF therapy. The present study confirmed the predictive value of symptom duration on visual prognosis and revealed 2 additional predictive factors: greater extent of hemorrhage and greater central foveal thickness. It has been shown that visual

prognosis is generally poor in cases in which the treatment of exudative AMD is delayed.26,27 Our result indicates that this is also true in cases of submacular hemorrhage. It also suggests that the nature of the causative lesion and the submacular hemorrhage severity influence visual prognosis, which highlights the need for prompt treatment. The extent of hemorrhage and central foveal thickness may represent the amount of subretinal hemorrhage. It is well known that subretinal hemorrhage itself induces retinal damage and degeneration.28,29 A thick hemorrhage also would limit the penetration of anti-VEGF to the exudative lesion. The close association of these 2 factors with visual acuity at 6 months may reflect this negative influence of subretinal hemorrhage on visual function. The visual prognosis was relatively poor in eyes with extensive hemorrhage or severe subfoveal hemorrhage. Considering the results of previous reports that the visual prognosis for eyes with subretinal hemorrhage is typically poor,2,3 our result may reflect the poor prognosis of the disease itself. However, the possibility that anti-VEGF monotherapy has limited efficacy in these cases cannot be completely neglected. That is, more aggressive treatment, such as pneumatic displacement or the use of tissue plasminogen activator, may be necessary in cases with large subretinal hemorrhage. In particular, these approaches will be considered in eyes exhibiting an extent of hemorrhage that is 9 disc areas or more or a central foveal thickness that is 704 mm or more at diagnosis because poor visual outcome will be suspected under either of these conditions. However, to our knowledge, the difference in efficacy between antiVEGF monotherapy and other treatment modalities in exudative AMD with submacular hemorrhage has not been investigated. A comparison of the efficacy of anti-VEGF monotherapy and other treatment modalities is beyond the scope of this study. We hope further studies will reveal the optimal treatment when a large amount of subretinal hemorrhage is present. In this study, symptom duration and extent of hemorrhage at baseline were not associated with the baseline BCVA. At diagnosis, one of the primary factors for determining visual acuity may be the amount of hemorrhage just beneath the fovea because the hemorrhage itself blocks visual stimulus to the fovea. However, the subfoveal hemorrhage often completely resolved or at least markedly decreased in the majority of eyes at 6 months, suggesting that the remaining retinal function itself is a primary factor for determining visual acuity at this time point. We propose that this difference in the primary determining factor of visual acuity between the 2 time points is the main reason why the symptom duration was associated with the BCVA at 6 months but not at diagnosis. In addition, the extent of hemorrhage was not associated with the BCVA at diagnosis. Moreover, the association between the extent of hemorrhage and the central foveal thickness at diagnosis was not significant in this study. This result suggests that the amount of hemorrhage beneath the fovea, which may be one of the primary determining factors of visual acuity, can be small even when the extent of the hemorrhage is large. This may partially explain the insignificant association between the extent of hemorrhage and the BCVA at diagnosis.

933

Ophthalmology Volume 121, Number 4, April 2014 In the present study, 8 eyes (8.8%) exhibited deteriorations in visual acuity of 3 lines or more, whereas previous studies reported deteriorations of no more than 2 lines.19e21 Three of these 8 eyes were diagnosed with typical exudative AMD and 3 with PCV. The lesion was not clearly diagnosed in the remaining 2 eyes. Although these 8 eyes exhibited slightly more extensive submacular hemorrhage (mean, 11.0 vs. 7.4 disc areas) and older mean age (73.4 vs. 69.6 years), the mean central foveal thickness (515.4 vs. 558.7 um) was slightly lower and the mean baseline BCVA (Snellen equivalent, 20/276 vs. 20/250) was slightly better than that in the other 83 eyes. Although we could not determine the reason for this marked deterioration in visual acuity, this result may highlight the importance of warning the patient that his or her visual acuity may deteriorate despite anti-VEGF therapy. The visual prognosis for PCV is generally superior to that for typical exudative AMD.30,31 In this study, however, visual acuity at 6 months was similar between both disease groups of patients. This result might originate from the greater extent of hemorrhage and greater central foveal thickness observed in those with PCV compared with those with typical exudative AMD. In this study, the central foveal thickness and the number of anti-VEGF injections were relatively comparable to those of the eyes that did not develop vitreous hemorrhage. However, the extent of vitreous hemorrhage was approximately 2 times greater than that of the eyes that did not develop vitreous hemorrhage. This result may suggest that the extent of hemorrhage is a factor predictive of the development of vitreous hemorrhage during the treatment. The relatively superior visual outcome in this study compared with previous observational studies suggests that anti-VEGF monotherapy has a valid effect. However, additional analysis showed that the BCVA had significantly deteriorated when compared with the BCVA that was measured before the development of the hemorrhage. Although the change in the BCVA was significantly worse in the hemorrhage during follow-up group compared with the hemorrhage at initial presentation group, a direct comparison between these 2 groups may not be appropriate for several reasons. First, the disease duration was longer in the hemorrhage during follow-up group; thus, the retinal function might have already been compromised in many of the eyes in the hemorrhage during follow-up group, which would strengthen the negative influence of the hemorrhage. Second, the BCVA before the development of surgery was not measured immediately before the development of the hemorrhage. A period of approximately 3 months passed between the time of measurement and the development of the hemorrhage, suggesting that further deterioration in visual function may have occurred during this period. Last, the baseline extent of the hemorrhage, which was determined to be one of the factors predictive of 6-month visual outcome, was relatively greater in the hemorrhage during follow-up group. The natural course and the treatment outcome of subretinal hemorrhages that develop during the follow-up of exudative AMD have not been fully elucidated. Thus, it is not clear whether the discouraging visual outcome in the hemorrhage during follow-up group

934

primarily originated from the devastating natural course of the disease or the limited efficacy of anti-VEGF monotherapy. An assessment of this association is beyond the scope of this study. Further studies will be needed to address these questions.

Study Limitations First, this study was retrospective, and the follow-up period was relatively short. Second, although our conclusions are based on the data for a relatively large study population, 42.8% of patients were excluded from the result analysis for various reasons. Because the extent of hemorrhage and central foveal thickness were greater among the excluded patients compared with the included patients, the outcomes also may have differed. In addition, as mentioned earlier, the difference in visual acuity outcome between typical exudative AMD and PCV was determined on the basis of the results of patients with definitive diagnoses (65/91 patients; 71.4%). Anti-VEGF injection was performed according to the discretion of each clinician and was not based on a single defined guideline; thus, the number of anti-VEGF injections was not controlled. Because a single guideline for selecting a treatment modality was lacking, the precise reason for selecting a specific treatment modality could not be accurately verified. In addition, the PCV cases may have been under-represented because we excluded cases that underwent photodynamic therapy. Last, anatomic outcomes could not be analyzed for all of the included patients. In conclusion, this 6-month follow-up study of 91 eyes with submacular hemorrhage secondary to exudative AMD showed that visual acuity in most patients improved or remained stable after anti-VEGF monotherapy. However, the marked deterioration in visual acuity observed in approximately 9% of eyes may suggest that this line of treatment has only limited efficacy in some cases. In addition, anti-VEGF therapy may not guarantee the restoration of visual function before the development of hemorrhage. Although eyes with PCV showed greater central foveal thickness, visual outcomes were similar among the various subsets of exudative AMD. The close associations between the visual acuity at 6 months after treatment and the duration of symptoms, extent of hemorrhage, and central foveal thickness at the time of diagnosis suggest that the exudative lesion itself and the extent of submacular hemorrhage may influence the visual prognosis. Future studies comparing the efficacy of antiVEGF monotherapy with other treatment modalities in patients with submacular hemorrhage would be of value.

References 1. Bennett SR, Folk JC, Blodi CF, Klugman M. Factors prognostic of visual outcome in patients with subretinal hemorrhage. Am J Ophthalmol 1990;109:33–7. 2. Avery RL, Fekrat S, Hawkins BS, Bressler NM. Natural history of subfoveal subretinal hemorrhage in age-related macular degeneration. Retina 1996;16:183–9. 3. Scupola A, Coscas G, Soubrane G, Balestrazzi E. Natural history of macular subretinal hemorrhage in age-related macular degeneration. Ophthalmologica 1999;213:97–102.

Kim et al



Treatment of Submacular Hemorrhage

4. Haupert CL, McCuen BW II, Jaffe GJ, et al. Pars plana vitrectomy, subretinal injection of tissue plasminogen activator, and fluid-gas exchange for displacement of thick submacular hemorrhage in age-related macular degeneration. Am J Ophthalmol 2001;131:208–15. 5. Kamei M, Tano Y, Maeno T, et al. Surgical removal of submacular hemorrhage using tissue plasminogen activator and perfluorocarbon liquid. Am J Ophthalmol 1996;121:267–75. 6. Vander JF, Federman JL, Greven C, et al. Surgical removal of massive subretinal hemorrhage associated with age-related macular degeneration. Ophthalmology 1991;98:23–7. 7. Submacular Surgery Trials (SST) Research Group. Surgery for hemorrhagic choroidal neovascular lesions of age-related macular degeneration: ophthalmic findings: SST report no. 13. Ophthalmology 2004;111:1993–2006. 8. Bakri SJ, Nickel J, Yoganathan P, Beer PM. Photodynamic therapy for choroidal neovascularization associated with submacular hemorrhage in age-related macular degeneration. Ophthalmic Surg Lasers Imaging 2006;37:278–83. 9. Ahmad S, Bearelly S, Stinnett SS, et al. Photodynamic therapy for predominantly hemorrhagic lesions in neovascular age-related macular degeneration. Am J Ophthalmol 2008;145:1052–7. 10. Hassan AS, Johnson MW, Schneiderman TE, et al. Management of submacular hemorrhage with intravitreous tissue plasminogen activator injection and pneumatic displacement. Ophthalmology 1999;106:1900–7. 11. Handwerger BA, Blodi BA, Chandra SR, et al. Treatment of submacular hemorrhage with low-dose intravitreal tissue plasminogen activator injection and pneumatic displacement. Arch Ophthalmol 2001;119:28–32. 12. Mizutani T, Yasukawa T, Ito Y, et al. Pneumatic displacement of submacular hemorrhage with or without tissue plasminogen activator. Graefes Arch Clin Exp Ophthalmol 2011;249:1153–7. 13. Ohji M, Saito Y, Hayashi A, et al. Pneumatic displacement of subretinal hemorrhage without tissue plasminogen activator. Arch Ophthalmol 1998;116:1326–32. 14. Rosenfeld PJ, Brown DM, Heier JS, et al; MARINA Study Group. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006;355:1419–31. 15. Avery RL, Pieramici DJ, Rabena MD, et al. Intravitreal bevacizumab (Avastin) for neovascular age-related macular degeneration. Ophthalmology 2006;113:363–72. 16. Spaide RF, Laud K, Fine HF, et al. Intravitreal bevacizumab treatment of choroidal neovascularization secondary to agerelated macular degeneration. Retina 2006;26:383–90. 17. Meyerle CB, Freund KB, Iturralde D, et al. Intravitreal bevacizumab (Avastin) for retinal angiomatous proliferation. Retina 2007;27:451–7. 18. Hikichi T, Higuchi M, Matsushita T, et al. One-year results of three monthly ranibizumab injections and as-needed

19.

20.

21. 22.

23.

24.

25.

26. 27.

28. 29. 30. 31.

reinjections for polypoidal choroidal vasculopathy in Japanese patients. Am J Ophthalmol 2012;154:117–24. Chang MA, Do DV, Bressler SB, et al. Prospective one-year study of ranibizumab for predominantly hemorrhagic choroidal neovascular lesions in age-related macular degeneration. Retina 2010;30:1171–6. Shienbaum G, Garcia Filho CA, Flynn HW Jr, et al. Management of submacular hemorrhage secondary to neovascular age-related macular degeneration with anti-vascular endothelial growth factor monotherapy. Am J Ophthalmol 2013;155:1009–13. Stifter E, Michels S, Prager F, et al. Intravitreal bevacizumab therapy for neovascular age-related macular degeneration with large submacular hemorrhage. Am J Ophthalmol 2007;144:886–92. Iacono P, Parodi MB, Introini U, et al. Intravitreal ranibizumab for choroidal neovascularization with large submacular hemorrhage in age-relatedmacular degeneration. Retina 2014;34: 281–7. Cho HJ, Koh KM, Kim HS, et al. Anti-vascular endothelial growth factor monotherapy in the treatment of submacular hemorrhage secondary to polypoidal choroidal vasculopathy. Am J Ophthalmol 2013;156:524–31. Krebs I, Schmetterer L, Boltz A, et al; MANTA Research Group. A randomised double-masked trial comparing the visual outcome after treatment with ranibizumab or bevacizumab in patients with neovascular age-related macular degeneration. Br J Ophthalmol 2013;97:266–71. Comparison of Age-related Macular Degeneration Treatments Trials (CATT) Research GroupMartin DF, Maguire MG, Fine SL, et al. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology 2012;119:1388–98. Arias L, Armada F, Donate J, et al. Delay in treating agerelated macular degeneration in Spain is associated with progressive vision loss. Eye (Lond) 2009;23:326–33. Oliver-Fernandez A, Bakal J, Segal S, et al. Progression of visual loss and time between initial assessment and treatment of wet age-related macular degeneration. Can J Ophthalmol 2005;40:313–9. Glatt H, Machemer R. Experimental subretinal hemorrhage in rabbits. Am J Ophthalmol 1982;94:762–73. Toth CA, Morse LS, Hjelmeland LM, Landers MB III. Fibrin directs early retinal damage after experimental subretinal hemorrhage. Arch Ophthalmol 1991;109:723–9. Yamashiro K, Tomita K, Tsujikawa A, et al. Factors associated with the response of age-related macular degeneration to intravitreal ranibizumab treatment. Am J Ophthalmol 2012;154:125–36. Laude A, Cackett PD, Vithana EN, et al. Polypoidal choroidal vasculopathy and neovascular age-related macular degeneration: same or different disease? Prog Retin Eye Res 2010;29:19–29.

Footnotes and Financial Disclosures Originally received: June 20, 2013. Final revision: September 17, 2013. Accepted: November 4, 2013. Available online: December 16, 2013.

Supported by Kim’s Eye Hospital Research Center. The sponsor or funding organization had no role in the design or conduct of this research. Manuscript no. 2013-990.

Department of Ophthalmology, Kim’s Eye Hospital, Konyang University College of Medicine, Seoul, South Korea.

Correspondence: Jae Hui Kim, MD, Department of Ophthalmology, Kim’s Eye Hospital, #156 Youngdeungpo-dong 4ga, Youngdeungpo-gu, Seoul, 150-034, South Korea. E-mail: [email protected].

Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

935