Comparison of Exudative Age-related Macular Degeneration Subtypes in Japanese and French Patients: Multicenter Diagnosis With Multimodal Imaging




Purpose


To compare and analyze differences and similarities between Japanese and French patients in subtype diagnosis of exudative age-related macular degeneration (AMD) as determined by fundus photography (FP) and fluorescein angiography (FA), and a multimodal imaging involving FP, FA, indocyanine green angiography (ICGA), and optical coherence tomography (OCT).


Design


Retrospective chart review.


Methods


We determined the subtype diagnosis for 99 consecutive Japanese eyes and 94 consecutive French eyes with exudative AMD. The first-step diagnosis was made using FP and FA, while the second-step diagnosis was made using FP, FA/ICGA, and OCT. The diagnoses made by Japanese and French physicians were compared, and when the diagnoses differed, a third institute was consulted to arrive at a final consensus and diagnosis.


Results


The first-step diagnosis showed 20%-30% disagreement against the final diagnosis, but the second-step diagnosis showed only 10% disagreement. Polypoidal choroidal vasculopathy (PCV) was observed more in Japanese patients (48%) than in French (9%), and the rate of PCV with type 1 or 2 choroidal neovascularization (CNV) was extremely low: 3% in Japanese and 0% in French. Type 1 CNV was found significantly more in French cases (53.3% vs 35.1%, P = .018), while the rate of eyes with type 2 CNV only or chorioretinal anastomosis was similar between populations.


Conclusions


Multimodality imaging significantly improved the sub-classification of AMD. There were significant differences between the 2 series in the proportions of type 1 CNV and PCV, while the proportions of type 2 CNV only and chorioretinal anastomosis were similar between groups.


Exudative age-related macular degeneration (AMD), one of the leading causes of visual impairment in developed countries, can be generally categorized clinically into 4 subtypes: (1) AMD with type 1 choroidal neovascularization (CNV), (2) AMD with type 1+2 CNV, (3) AMD with type 2 CNV only, and (4) chorioretinal anastomosis. An additional subtype (5) related to polypoidal choroidal vasculopathy (PCV), either (5a) without CNV or (5b) associated with type 1 or 2 CNV, has been suggested. Previous reports show ethnically different distributions for these subtypes. Differentiating these subtypes is important because the clinical course and response to treatments may differ depending on the subtype. Photodynamic therapy seems to be efficient for achieving the regression of polypoidal lesions in PCV, while anti–vascular endothelial growth factor (VEGF) treatment is more useful for regression of subretinal or intaretinal fluid. Photodynamic therapy combined with anti-VEGF treatment is recommended for chorioretinal anastomosis because chorioretinal anastomosis may be more resistant to anti-VEGF monotherapy. Interestingly, some reports have shown that anti-VEGF monotherapy works equally as well for PCV and chorioretinal anastomosis as for typical exudative AMD, although the classification of exudative AMD in these reports is not consistent or comparable.


A combined approach using fundus photography and fluorescein angiography (FA) has been the gold standard in the diagnosis and treatment of exudative AMD. However, in some cases, fundus photography and FA alone are insufficient for the final diagnosis of various subtypes of exudative AMD. For example, some PCV lesions can appear similar to “classic” type 2 CNV in FA images. Considering that the pathophysiology of PCV and chorioretinal anastomosis has been established, in large part owing to recent advances in indocyanine green angiography (ICGA) and optical coherence tomography (OCT) devices, the use of multimodality imaging diagnostic devices should be helpful in making more accurate subtype diagnoses of exudative AMD. However, there are sometimes cases in which a clear distinction among the 4 subtypes of exudative AMD and/or the 2 subtypes of PCV cannot be made, even with modern multimodality diagnostic devices. Moreover, polypoidal lesions can develop in eyes diagnosed with typical exudative AMD, mainly those with late type 1 or occult CNV.


Besides differences in imaging, the training and experience of physicians in different countries can also affect the subtype diagnosis of exudative AMD. Asian physicians may be more familiar with PCV but less so with type 1 or occult CNV and chorioretinal anastomosis, while Western physicians may be more familiar with type 1 or occult CNV and chorioretinal anastomosis, but less so with typical PCV. However, most previous studies on the subtype distribution of exudative AMD have been conducted in individual facilities, and owing to lack of the standard protocol, this may have introduced biases.


To address these gaps in the literature, we conducted a comparative study in Japan and France to determine (1) whether subtype diagnosis of exudative AMD can be made by fundus photography and FA alone, (2) whether the current multimodality diagnostic devices (fundus photography, FA, ICGA, and OCT) can further help to enhance the accuracy of subtype diagnosis, (3) whether all exudative AMD can be categorized into 4 subtypes of typical exudative AMD and 2 subtypes of PCV, and finally, (4) what is the distribution of AMD subtypes diagnosed with multimodal imaging according to ethnic differences at a French center as compared to that at a Japanese center.


Methods


We retrospectively reviewed the medical records of 99 eyes of 99 consecutive patients who visited the Department of Ophthalmology, Kyoto University Hospital with a tentative diagnosis of neovascular AMD (Kyoto cases) and 95 eyes of 85 patients with presumed neovascular AMD at Centre d’Ophtalmologie de Paris (Paris cases). All patients underwent comprehensive ophthalmic examinations, including the measurement of best-corrected visual acuity, intraocular pressure testing, indirect ophthalmoscopy, slit-lamp biomicroscopy with a contact lens, spectral-domain OCT (Spectralis HRA+OCT; Heidelberg Engineering, Heidelberg, Germany), and FA/ICGA (HRA-2; Heidelberg Engineering). All procedures in this study adhered to the tenets of the Declaration of Helsinki. The Kyoto University ethics committee and Paris University ethics committee approved the retrospective research on their examinations for this study.


At Kyoto University, 2 retina specialists (K.Y. and A.T.) evaluated fundus photography and FA and made the “first-step diagnosis” for both Kyoto cases and Paris cases. If the specialists disagreed regarding the diagnosis, a third retina specialist (N.Y.) was consulted for the final determination. Multimodal images of fundus photography, FA, ICGA, and OCT results were used to make a “second-step diagnosis.”


At Centre d’Ophtalmologie de Paris, 2 retina specialists (B.D.U. and F.C.) evaluated fundus photography and FA for the “first-step diagnosis” and multimodal images of fundus photography, FA, ICGA, and OCT assessments were used to make a “second-step diagnosis.” In the case of disagreement, a third retina specialist (G.C.) determined the diagnosis.


When the “second-step diagnosis” made by the 2 institutes agreed, the diagnosis was regarded as the “final diagnosis.” When the diagnosis by the 2 institutes failed to reach a consensus, retina specialists (C.M.G.C. and T.Y.W.) at Singapore Eye Research Institute were consulted for a diagnosis. In such cases, the diagnosis by Singapore Eye Research Institute was regarded as the “final diagnosis.” Singapore Eye Research Institute investigators were asked to use all the data and also to choose 1 from the 2 “second-step diagnoses” suggested previously. In each case, Singapore Eye Research Institute investigators agreed on 1 of the “second-step diagnoses” provided by the Kyoto and Paris investigators. Both Kyoto and Paris cases were subgrouped into: (1) AMD with type 1 CNV; (2) AMD with type 1+2 CNV; (3) AMD with type 2 CNV only; and (4) chorioretinal anastomosis. An additional type (5) is caused by PCV, either (5a) without CNV or (5b) associated with type 1 or 2 CNV. Eyes with PCV with branching vascular network without CNV were categorized to (5a) PCV without CNV.


A diagnosis of PCV was made based on fundus photography, FA/ICGA, and OCT: elevated orange-red lesions, characteristic polypoidal lesions at the edge of a branching vascular network on angiography, and prominent anterior protrusion of the retinal pigment epithelium line in OCT images. A diagnosis of chorioretinal anastomosis was also made based on fundus photography, FA/ICGA, and OCT: subretinal, intraretinal, or preretinal juxtafoveal hemorrhages; dilated retinal vessels; lipid exudates; and retinal–choroidal anastomosis. For the analysis of AMD subtypes, AMD with type 1 CNV, AMD with type 2 CNV, and AMD with type 1+2 CNV were regarded as typical exudative AMD, and PCV associated with type 1 or 2 CNV and PCV without type 1 or 2 CNV were regarded as PCV. Fisher exact test was used to compare groups. P values of less than .05 were considered to be statistically significant.




Results


This study consisted of 99 consecutive eyes of 99 Japanese patients and 95 consecutive eyes of 85 French patients with presumed neovascular AMD. One eye from among the French was excluded from the study because of angiographic images of low quality. The mean age of the 99 Japanese patients (70 men and 29 women) was 74.0 ± 8.9 years, and all patients were ethnically Japanese. The mean age of the 85 French patients (45 men and 40 women) was 73.5 ± 7.9 years, and 98% were white.


The “first-step diagnosis” for each of the 99 Kyoto cases, made by Kyoto investigators using fundus photography and FA, is shown in Table 1 . Among these “first-step diagnoses,” 34.3% (34/99) differed from the “final diagnosis” as determined by the 3 facilities together. The number of eyes for which the diagnosis involved disagreement decreased to 10 (10.1%) when considering the “second-step diagnosis,” which was based on the additional information provided by ICGA and OCT. Figure 1 shows the breakdown of eyes for which diagnosis differed from the “final diagnosis.”



Table 1

Two-Step Diagnosis of Neovascular Age-related Macular Degeneration Subtypes in Kyoto Cases (N = 99)































































By Kyoto Investigators By Paris Investigators Final Diagnosis
First-Step Diagnosis Second-Step Diagnosis First-Step Diagnosis Second-Step Diagnosis
AMD with type 1 CNV 46 (46.5%) 34 (34.3%) 34 (34.3%) 28 (28.3%) 33 (33.3%)
AMD with type 1+2 CNV 17 (17.2%) 8 (8.1%) 1 (1.0%) 3 (3.0%) 6 (6.1%)
AMD with type 2 CNV 2 (2.0%) 3 (3.0%) 6 (6.1%) 7 (7.1%) 5 (5.1%)
Chorioretinal anastomosis 4 (4.0%) 5 (5.1%) 4 (4.0%) 4 (4.0%) 5 (5.1%)
PCV with type 1 or 2 CNV 1 (1.0%) 3 (3.0%) 2 (2.0%) 5 (5.1%) 3 (3.0%)
PCV without type 1 or 2 CNV 24 (24.2%) 41 (41.0%) 48 (48.5%) 47 (47.5%) 42 (42.4%)
Other 5 (5.1%) 5 (5.1%) 4 (4.0%) 5 (5.1%) 5 (5.1%)

AMD = age-related macular degeneration; CNV = choroidal neovascularization; PCV = polypoidal choroidal vasculopathy.

First-step diagnosis was determined using fundus photography and fluorescein angiography, while second-step diagnosis was determined using fundus photography, fluorescein angiography, indocyanine green angiography, and optical coherence tomography.



Figure 1


Breakdown of eyes with disagreements when “first-step” and “final diagnosis” were compared in Kyoto cases of neovascular age-related macular degeneration. First-step diagnosis was determined using fundus photography and fluorescein angiography, while second-step diagnosis was determined using fundus photography, fluorescein angiography, indocyanine green angiography, and optical coherence tomography. Circles in the left column show the frequency of each diagnosis among Kyoto investigators, while circles in the right column show the frequency of each diagnosis among Paris investigators. The inner circle shows the “first-step diagnosis,” and the outer circle shows the “second-step diagnosis.” The squares in the middle column show the “final diagnosis” as determined by 3 facilities: Kyoto University, University Paris, and Singapore Eye Research Institute.


Most eyes for which diagnosis involved disagreement were diagnosed as having type 1 CNV, AMD with type 1 CNV, or AMD with type 1+2 CNV. Notably, despite the higher prevalence of PCV in Japan, Kyoto investigators are rather reluctant to make a diagnosis of PCV with fundus photography and FA alone, and prefer to diagnose eyes with PCV as having AMD with type 1 CNV; 44.9% (22/49) of eyes with PCV with/without CNV were initially diagnosed with AMD with type 1 CNV rather than PCV. However, in the “second-step diagnosis” with fundus photography, FA/ICGA, and OCT, only 3 eyes were judged as AMD; the other 93.9% (46/49) of cases were diagnosed as PCV. Figure 2 shows an example of eyes diagnosed as having AMD with type 1 CNV in the “first-step diagnosis” but as PCV in the “second-step diagnosis.”




Figure 2


Right eye of an 81-year-old man with age-related macular degeneration among the Kyoto cases; an example of eyes with different diagnosis between the “first-step diagnosis” determined by fundus photography and fluorescein angiography and the “second-step diagnosis” determined by fundus photography, fluorescein angiography, indocyanine green angiography, and optical coherence tomography. Fundus photography (Top left) shows depigmentation of the retinal pigment epithelium at the nasal-lower area of the macula, suggesting the existence of choroidal neovascularization. Fluorescein angiography shows the faint leakage of fluorescein at the nasal-lower area of the macula in the early phase (Top middle) and substantial leakage in the late phase (Middle middle). The “first-step diagnosis” was age-related macular degeneration with type 1 CNV. Indocyanine green angiography shows polypoidal lesions at the edge of the subfoveal pigment epithelium detachment in the early phase (Top right). The polypoidal lesions and network vessels between the polypoidal lesions and the disc are clearly visible in the late phase (Middle right). Optical coherence tomography (Bottom) shows protrusions of the retinal pigment epithelium line indicating serous pigment epithelium detachment and a polypoidal lesion at its edge. The flat elevation of the retinal pigment epithelium line next to the polypoidal lesion indicates network vessels. The “second-step diagnosis” was polypoidal choroidal vasculopathy without type 1 or type 2 choroidal neovascularization.


The “first- and second-step diagnoses” of the Kyoto cases made by the Paris investigators showed different trends. The Paris physicians tended to make a diagnosis of more AMD with type 2 CNV, less AMD with type 1+2 CNV, and more PCV without CNV ( Table 1 ). The “first-step diagnosis” showed disagreement compared to the “final diagnosis” in 18.2% (18/99) of cases, while the “second-step diagnosis” showed disagreement compared to the “final diagnosis” in 13.1% (13/99) of cases. Although most eyes with PCV were correctly diagnosed using fundus photography and FA alone, approximately 20% (8/46 and 10/46, respectively) of eyes with type 1 CNV or chorioretinal anastomosis were diagnosed as having PCV at the time of both “first- and second-step diagnoses” by the Paris investigators ( Figure 1 ).


The “first-step diagnoses” for the 94 Paris cases made by the Paris investigators using fundus photography and FA are shown in Table 2 . Among these “first-step diagnoses,” 24.5% (23/94) differed from the “final diagnosis” as determined by the 3 facilities together. The number of eyes with any disagreement related to diagnosis decreased to 9 (9.6%) for the “second-step diagnosis” based on the additional information provided by ICGA and OCT. The rate of any disagreement related to diagnosis by Kyoto investigators similarly decreased from 20.2% (19/94) for the “first-step diagnosis” to 9.6% (9/94) for the “second-step diagnosis.”



Table 2

Two-step Diagnosis of Neovascular Age-related Macular Degeneration Subtypes in Paris Cases (N = 94)































































By Kyoto Investigators By Paris Investigators Final Diagnosis
First-Step Diagnosis Second-Step Diagnosis First-Step Diagnosis Second-Step Diagnosis
AMD with type 1 CNV 50 (53.2%) 52 (55.3%) 50 (53.2%) 50 (53.2%) 48 (51.1%)
AMD with type 1+2 CNV 25 (26.6%) 20 (21.3%) 12 (12.8%) 14 (14.9%) 19 (20.2%)
AMD with type 2 CNV 6 (6.4%) 6 (6.4%) 16 (17.0%) 13 (13.8%) 10 (10.6%)
Chorioretinal anastomosis 4 (4.3%) 5 (5.3%) 5 (5.3%) 5 (5.3%) 5 (5.3%)
PCV with type 1 or 2 CNV 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
PCV without type 1 or 2 CNV 6 (6.4%) 7 (7.4%) 3 (3.2%) 8 (9.5%) 8 (9.5%)
Other 3 (3.2%) 4 (4.3%) 7 (7.4%) 4 (4.3%) 4 (4.3%)

AMD = age-related macular degeneration; CNV = choroidal neovascularization; PCV = polypoidal choroidal vasculopathy.

The first-step diagnosis was determined using fundus photography and fluorescein angiography, while the second-step diagnosis was determined using fundus photography, fluorescein angiography, indocyanine green angiography, and optical coherence tomography.


Figure 3 shows the breakdown of eyes for which there was any disagreement related to the “final diagnosis.” For the Paris investigators, most disagreements were associated with eyes with a “final diagnosis” of AMD with type 1+2 CNV, while disagreements related to diagnosis among Kyoto investigators were often observed in eyes with type 2 CNV. The diagnoses made by Paris investigators showed more AMD with type 2 CNV and less AMD with type 1+2 CNV in comparison to the diagnoses made by Kyoto investigators. Figure 4 shows an example of eyes diagnosed as having type 1 CNV in the “first-step diagnosis” but as having type 2 CNV in the “second-step diagnosis.”




Figure 3


Breakdown of eyes with disagreement between “first-step” and “final diagnosis” among the Paris cases of neovascular age-related macular degeneration. First-step diagnosis was determined using fundus photography and fluorescein angiography, while second-step diagnosis was determined using fundus photography, fluorescein angiography, indocyanine green angiography, and optical coherence tomography. Circles in the left column show the frequency of each diagnosis determined by Kyoto investigators, while circles in the right column show the frequency of each diagnosis determined by Paris investigators. The inner circle shows the diagnosis in the “first-step diagnosis,” and the outer circle shows the diagnosis in the “second step diagnosis.” The squares in the middle column show the “final diagnosis” determined by 3 facilities: Kyoto University, University Paris, and Singapore Eye Research Institute.

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 Comparison of Exudative Age-related Macular Degeneration Subtypes in Japanese and French Patients: Multicenter Diagnosis With Multimodal Imaging

Full access? Get Clinical Tree

Get Clinical Tree app for offline access