Subfoveal Choroidal Thickness as a Potential Predictor of Visual Outcome and Treatment Response After Intravitreal Ranibizumab Injections for Typical Exudative Age-Related Macular Degeneration


To investigate the prognostic implication of subfoveal choroidal thickness on treatment outcome after intravitreal ranibizumab injections for typical exudative age-related macular degeneration (AMD).


Retrospective study.


A total of 40 eyes of 37 patients who completed 6-month follow-up were analyzed. Patients’ data were retrieved from medical records including best-corrected visual acuity (BCVA). Subfoveal choroidal thickness at baseline, 3 months, and 6 months was measured by enhanced depth imaging optical coherence tomography and adjusted for age and sex before statistical analysis. Treatment response was after 3 monthly intravitreal ranibizumab injections. Responders (responder group) were defined as a 100 μm or more decrease or complete resolution of subretinal fluid, whereas nonresponders (nonresponder group) were defined as changes less than 100 μm or more than 100 μm increase of subretinal fluid by optical coherence tomography.


Mean age at diagnosis was 72.1 ± 8.1 years, and 22 eyes (55.0%) were responders. The responder group had thicker subfoveal choroid (257.2 ± 108.3 μm) and smaller lesions (1.3 ± 0.8 μm) at baseline than the nonresponder group (167.1 ± 62.4 μm, P = .003; and 2.0 ± 1.0 μm, P = .008). The responder group showed significantly better BCVA and thicker subfoveal choroid than the nonresponder group at 3 months ( P = .002 and P = .023) and 6 months ( P = .004 and P = .031). Stepwise and binary regression analysis demonstrated that subfoveal choroidal thickness was significantly correlated with visual outcome (B = −0.002, P = .003) and treatment response (B = 8.136, P = .018).


Subfoveal choroidal thickness may be a predictive factor for visual outcome and treatment response in typical exudative AMD after intravitreal ranibizumab injections.

Choroidal circulation provides nutrients and removes metabolic wastes from retinal pigment epithelium (RPE) and the outer retina, playing an important role in maintaining normal visual function. Thus, impaired choroidal circulation may disrupt normal retinal function, subsequently leading to visual deterioration. It has been well documented that choroidal flow and the volume of the choriocapillaris are negatively correlated with aging.

Age-related macular degeneration (AMD) is a leading cause of visual loss in developed countries and is a multifactorial disease related to age, genetics, and environmental factors such as smoking. Based on several studies, abnormalities of choroidal circulation have been suggested as one of the factors involved in the pathogenesis of AMD. Some studies suggested that inadequate choroidal perfusion or ischemia within the choroidal microarchitecture could lead to hypoxia and ischemia of the overlying RPE. In addition, eyes with dry AMD showed decreased blood volume and abnormal choroidal flow, which were inversely related with disease severity, such as drusen severity.

The introduction of enhanced depth imaging optical coherence tomography (EDI-OCT) has provided better choroidal signal penetration by using increased wavelengths. In addition to the evidence, which suggests the dynamic role of choroidal circulation in the pathogenesis of AMD, the EDI-OCT devices have provided additional information on choroidal thickness in AMD. It has been shown that the patients with AMD have thinner choroidal thickness than age-matched normal controls. In dry AMD, drusen load was inversely correlated with choroidal thickness. In addition, EDI-OCT has shown distinct characteristics in variants of AMD. The eyes with retinal angiomatous proliferation had a thinner choroid, whereas those with polypoidal choroidal vasculopathy had a thicker choroid than those with typical exudative AMD.

Although the impact of choroidal circulation and its thickness have been implicated in the pathogenesis of AMD, the relationship of choroidal thickness with visual outcome and treatment response after anti–vascular endothelial growth factor (VEGF) therapy has not been well documented in patients with typical exudative AMD. Herein, we evaluated the subfoveal choroidal thickness of eyes with typical exudative AMD and its association with treatment response after intravitreal ranibizumab injections during the follow-up of 6 months. In addition, we investigated the prognostic implication of subfoveal choroidal thickness in both treatment response and visual outcome in typical exudative AMD patients.


Enrollment of Study Subjects

We retrospectively reviewed the medical records of patients with treatment-naïve exudative AMD who were treated at the Vitreoretinal Service Clinic of Yonsei University Medical Center from March 1, 2011 to March 31, 2012. All patients were treated with 3 monthly intravitreal ranibizumab injections and then as-needed intravitreal ranibizumab injections. The patients completed at least a 6-month follow-up after the first intravitreal ranibizumab injections. This retrospective study was approved by the Institutional Review Board of Yonsei University College of Medicine, and was conducted in accordance with the tenets of the Declaration of Helsinki. A diagnosis of exudative AMD was based on the color fundus photography, fluorescein angiography (FA), and indocyanine green angiography (ICGA) results of each patient, with evidence of hyperfluorescence and late leakage associated with at least 1 of the following signs: subretinal membrane, exudative subretinal fluid, or subretinal hemorrhage involving the fovea. All eyes compatible with polypoidal choroidal vasculopathy, showing branching vascular networks with terminating polypoidal lesions on ICGA, were excluded. In addition, eyes with retinal angiomatous proliferation that exhibited retinal-retinal or retinal-choroidal anastomosis were also excluded. Further excluded were patients with: (1) any previous treatments for exudative AMD, such as laser photocoagulation, photodynamic therapy, or any intravitreal injections; (2) eyes with spherical equivalent refractive error of more than ±3.0 diopters; (3) no evidence of subfoveal involvement; (4) any sign of advanced AMD, such as disciform scarring, to minimize possible effect of disease duration; (5) concomitant ocular diseases that might further compromise visual acuity, such as epiretinal membrane or diabetic retinopathy; and (6) patients with aphakia, absence of posterior capsule, or history of vitrectomy attributable to reduced ranibizumab retention.


All patients underwent complete ocular examination, including best-corrected visual acuity (BCVA) measured using decimal visual acuity charts, slit-lamp examination, and dilated funduscopic examination via indirect ophthalmoscopy. Color fundus photography and OCT images were also obtained. FA and ICGA images were obtained by use of a Heidelberg retinal angiograph (HRA-II; Heidelberg Engineering, Heidelberg, Germany). Choroidal neovascularization (CNV) attributable to AMD was classified as type 1 and type 2 CNV based on FA findings. Type 1 occult CNV was defined as fibrovascular pigment epithelial detachment or late-phase leakage of an undetermined source. Type 2 classic CNV was defined by the typical characterization of an area of choroidal hyperfluorescence with well-demarcated boundaries that could be discerned in the early phase by FA. The cases of typical exudative AMD were then classified into 3 categories: predominantly classic (50%–100% of the lesion was classic), minimally classic (1%–49% of the lesion was classic), and occult type (100% of the lesion was composed of occult-only leakage). The locations of CNV lesions were subdivided into 3 categories: subfoveal, juxtafoveal (within 1–199 μm from the foveal center), and extrafoveal (200 μm or more distance from the foveal center). Using digital calipers provided by HRA-II software, the greatest linear diameter of CNV at baseline was measured manually on late-phase images of FA.

OCT images were obtained by spectral-domain OCT (Spectralis; Heidelberg Engineering). Central macular thickness (CMT) was defined as a mean thickness of the central 1-mm zone of the ETDRS grid and measured by the program embedded in spectral-domain OCT. In addition, OCT characteristics, including subretinal detachment, pigment epithelial detachment, and vitreomacular traction, were also analyzed.

The method of obtaining EDI-OCT has been described previously. EDI-OCT imaging was performed by positioning the objective lens of the Spectralis OCT scanner close enough to invert the image. At least 2 horizontal and vertical scans across the fovea for each eye were obtained with good quality. The OCT images were saved after 100 frames were averaged using the automatic averaging and eye tracking system of Spectralis OCT. Choroidal thickness was defined as the distance from the outer border of the hyper-reflective line, corresponding to the RPE perpendicular to the chorioscleral interface. Using digital calipers provided by the Heidelberg Spectralis OCT software, choroidal thickness was measured at the subfoveal region in both horizontal and vertical images and then averaged. Two independent observers who were masked to the clinical data of each patient (H.M.K. and H.J.K.) measured subfoveal choroidal thickness.

Follow-up visits were performed at 1 month after each monthly intravitreal ranibizumab injection. Follow-up visits included BCVA measurement, dilated funduscopic examination via indirect ophthalmoscopy, and OCT. After 3 monthly loading intravitreal ranibizumab injections, additional FA and ICGA examinations were performed.

Protocols for Intravitreal Ranibizumab Injections

Ranibizumab (Lucentis; Genentech Inc, South San Francisco, California, USA) was intravitreously injected in the operating room under strict aseptic conditions. A total of 0.05 mL of ranibizumab (0.05 mg of Lucentis) was injected into the vitreous cavity through the superior sclera using a 30 gauge needle, at a position 3.5 mm posterior to the corneal limbus in phakic eyes and 3.0 mm posterior in pseudophakic eyes. Pressure was applied to the injection site using a sterile cotton swab for 1 minute to prevent leakage. After injections, all patients were instructed to apply antibiotic eye drops 4 times a day for 1 week.

After 3 monthly loading injections of intravitreous ranibizumab, retreatment was considered if any of the following changes were observed after 3 monthly loading injections of intravitreous ranibizumab: (1) any visual acuity loss determined with a decimal visual acuity chart with OCT evidence of fluid in the macula; (2) an increased CMT of at least 100 μm; and (3) evidence of persistent fluid on OCT 1 month after the previous injection. When reinjection was required, ranibizumab was intravitreally injected within 1 week.

Definition of Treatment Response

Treatment response after 3 monthly intravitreal ranibizumab injections was determined anatomically by OCT, modifying the reinjection criteria of the PrONTO trial, and classified into 2 categories. The height of subretinal fluid was assessed on the macular region, and the highest spot of subretinal fluid was used for serial comparison. The “responder” was defined as resolution or more than 100 μm decrease of subretinal fluid at 3 months when compared with baseline examination. The “nonresponder” was defined as no significant change of subretinal fluid within 100 μm or increase of more than 100 μm of subretinal fluid at 3 months when compared with baseline examination.

Statistical Analysis

Patients’ characteristics were retrieved from the medical charts, including the age at diagnosis of exudative AMD, sex, type of CNV (predominantly classic, minimally classic, or occult), location of CNV (subfoveal, juxtafoveal, or extrafoveal), CMT at baseline OCT, and BCVA at baseline and at each follow-up. Decimal BCVA results were converted to a logarithm of the minimal angle of resolution (logMAR) value for statistical analysis. Upon statistical analysis, subfoveal choroidal thickness was adjusted for both sex and age.

We compared baseline characteristics between the “responders” (responder group) and “nonresponders” (nonresponder group). For subgroup analysis, nonparametric analyses were used: Mann-Whitney U test for continuous variables and χ 2 test for categorical variables.

IBM SPSS Statistics 18.0 software for Windows (IBM Corporation, Somers, New York, USA) was used for the statistical analyses. To determine visual prognostic factors after intravitreal ranibizumab injections for typical exudative AMD, possible factors were analyzed using stepwise regression analysis. Included factors were age, sex, BCVA at baseline, the greatest linear diameter of CNV at baseline, characteristics of baseline OCT images including subfoveal choroidal thickness and CMT at baseline, and type and location of CNV attributable to AMD. Binary forward conditional regression analysis was used for investigation of predictive factors for response after 3 monthly intravitreal ranibizumab injections. Paired t tests were used to compare both mean BCVA and subfoveal choroidal thickness at each time point, with the corresponding baseline values. Mauchly’s test of sphericity and Kolmogorov-Smirnov analyses were used to confirm statistical validity. Results with P < .05 were considered statistically significant.


Baseline Characteristics

In total, 40 eyes of 37 patients were included and analyzed in this retrospective study. The patients had a mean age of 72.1 ± 8.1 years (range 52–86 years), and mean spherical equivalent refractive error was 0.3 ± 1.3 diopters (range −2.4 diopters to +2.5 diopters). Baseline FA showed predominantly classic CNV in 14 eyes (35.0%), minimally classic in 10 eyes (25.0%), and occult CNV in 16 eyes (40.0%). The patients’ clinical details are listed in Table 1 .

Table 1

Baseline Characteristics of Study Patients With Typical Exudative Age-Related Macular Degeneration

Eyes/patients 40/37
Mean age at diagnosis (y) 72.1 ± 8.1 a (range, 52–86)
Male/female 24 (64.9%)/13 (35.1%)
Right eye/left eye 20 (50.5%)/20 (50.5%)
Mean spherical equivalent (diopters) 0.3 ± 1.3 (range, −2.4 to +2.5)
logMAR 0.85 ± 0.53 a
(Snellen equivalent) (20/141)
Mean central macular thickness (μm) 474.6 ± 324.8 a (range, 256.0–7380)
Mean subfoveal choroidal thickness (μm) 216.7 ± 100.4 (range, 76.0–441.0)
Mean greatest linear diameter of CNV (mm) 1.6 ± 1.1 (range, 0.5–4.7)
Type of CNV, eyes (%)
Predominantly classic 14 (35.0%)
Minimally classic 10 (25.0%)
Occult 16 (40.0%)
Location of CNV, eyes (%)
Subfovea 27 (67.5%)
Juxtafovea 7 (17.5%)
Extrafovea 6 (15.0%)
OCT characteristics, eyes (%)
Intraretinal fluid cyst 18 (45.0%)
Vitreomacular traction 5 (12.5%)
Pigment epithelial detachment 15 (37.5%)
Subretinal fluid 32 (80.0%)

BCVA = best-corrected visual acuity; CNV = choroidal neovascularization; logMAR = logarithm of the minimal angle of resolution; OCT = optical coherence tomography.

a Mean ± standard deviation.

Comparison Between the Responders and Nonresponders

After 3 monthly loading injections of ranibizumab intravitreously, we classified the eyes into responders and nonresponders. Twenty-two of 40 eyes (55.0%) were classified as responders and 18 eyes (45.0%) were nonresponders by OCT analysis. Among 3 patients with bilateral AMD, one patient had bilateral ‘responder’ eyes, and another patient had bilateral ‘nonresponder’ eyes. The other patient had one responder eye and one nonresponder eye. In the responder group, 13 eyes (59.1%) showed complete resolution of subretinal fluid on OCT after 3 monthly loading injections. Seven eyes (40.9%) with decreased subretinal fluid underwent further treatment, and mean 1.0 ± 1.2 intravitreal ranibizumab injections were done additionally. All eyes in the nonresponder group underwent additional injections. In total, intravitreal ranibizumab injections were done 4.0 ± 1.2 times in the responder group and 5.8 ± 0.7 times in the nonresponder group ( P < .001).

We compared characteristics between the responder group and the nonresponder group. Among the baseline characteristics, mean subfoveal choroidal thickness and CNV size were significantly different between the 2 groups. Mean subretinal choroidal thickness of the responder group was significantly thicker than that of the nonresponder group ( P = .003). In addition, mean baseline CNV size was significantly smaller in the responder group than in the nonresponder group ( P = .008). The comparison of baseline characteristics between the 2 groups is shown in Table 2 .

Table 2

Comparison of Baseline Characteristics Between the Responders (Responder Group) and Nonresponders (Nonresponder Group) After 3 Monthly Intravitreal Ranibizumab Injections for Typical Exudative Age-Related Macular Degeneration

Responder Group (n = 22 eyes) Nonresponder Group (n = 18 eyes) P Value
Mean age at diagnosis (y) 70.1 ± 9.4 74.5 ± 5.4 .090 a
Mean spherical equivalent (diopters) 0.3 ± 1.1 0.4 ± 1.6 .813 a
logMAR 0.77 ± 0.47 0.94 ± 0.59 .319 a
(Snellen equivalent) (20/117) (20/174)
Mean central macular thickness (μm) 423.2 ± 292.1 537.4 ± 359.3 .274 a
Mean subfoveal choroidal thickness (μm) 257.2 ± 108.3 167.1 ± 62.4 .003 a
Mean greatest linear dimension of CNV (mm) 1.3 ± 0.8 2.0 ± 1.0 .008 a
Type of CNV .273 b
Predominantly classic 10 (45.5%) 4 (22.3%)
Minimally classic 4 (18.1%) 6 (33.3%)
Occult 8 (36.4%) 8 (44.4%)
Location of CNV .141 b
Subfovea 12 (54.6%) 15 (83.2%)
Juxtafovea 5 (22.7%) 2 (11.2)
Extrafovea 5 (22.7%) 1 (5.6%)
OCT characteristics
Intraretinal fluid cyst 9 (40.9%) 9 (50.0%) .577 b
Vitreomacular traction 3 (13.6%) 2 (11.1%) .598 b
Pigment epithelial detachment 5 (22.7%) 10 (55.6%) .051 b
Subretinal fluid 16 (72.7%) 16 (88.9%) .214 b

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 8, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Subfoveal Choroidal Thickness as a Potential Predictor of Visual Outcome and Treatment Response After Intravitreal Ranibizumab Injections for Typical Exudative Age-Related Macular Degeneration

Full access? Get Clinical Tree

Get Clinical Tree app for offline access