To evaluate the efficacy of spectral-domain optical coherence tomography (SD OCT) compared to indocyanine green angiography (ICGA) in detecting idiopathic polypoidal choroidal vasculopathy (PCV) and in differentiating between PCV and occult choroidal neovascularization (CNV).
Retrospective observational case-control study.
SD OCTs of 51 eyes of 44 consecutive patients who presented with 1 or more pigment epithelial detachments (PEDs) attributable to either PCV or occult CNV were retrospectively reviewed by a grader masked to the final diagnosis. A qualitative analysis based on the following tomographic findings was performed: sharp PED peak, PED notch, hyporeflective lumen within hyperreflective lesions adherent to retinal pigment epithelium. The diagnosis based on SD OCT alone was compared with the final diagnosis made using ICGA and fluorescein angiography. Sensitivity and specificity were calculated. Patients with classic CNV and central serous chorioretinopathy were excluded.
Among 51 eyes of 44 patients, 37 had an ICGA-confirmed diagnosis of PCV and 14 had occult CNV. SD OCT based on the features above detected 35 of 37 true-positive PCV lesions but missed 2 ICGA-confirmed lesions (false negatives). SD OCT correctly excluded 13 of 14 non-PCV lesions but misidentified 1 PCV lesion (false positive). These data showed a sensitivity of 94.6% and a specificity of 92.9% for the above SD OCT features in identifying PCV lesions.
SD OCT based on the features above allowed for good detection of PCV and differentiation between PCV and occult CNV in this selected clinic population. A careful qualitative analysis of the tomographic findings in patients presenting with PEDs may allow ophthalmologists to distinguish between PCV and occult CNV, decreasing the need for ICGA and the risks related to this procedure.
Idiopathic polypoidal choroidal vasculopathy (PCV) is an abnormality of the choroidal vasculature first reported by Yannuzzi and associates. They described the condition as “vascular nodular protrusions emanating from the choroid,” which clinically presents with serous and hemorrhagic detachment of the neurosensory retina and pigment epithelium, both at the macula and at the peripapillary area. At present, indocyanine green angiography (ICGA) represents the modality of choice to best visualize and diagnose the condition. In fact, 2 distinguishing choroidal vascular changes have been described on ICGA: a branching network of vessels in the inner choroid, and vascular dilations at the border of the network of vessels. With the advent and utilization of optical coherence tomography (OCT), which can provide close to histologic visualization of the retina, a number of groups have used this modality to study the etiology and pathogenesis of PCV. In particular, spectral-domain OCT (SD OCT) is proving to be a very useful tool to describe new anatomic/tomographic features of PCV. Nonetheless, the pathogenesis of PCV remains poorly understood. Further, ICGA is not widely available in many centers, leading to some uncertainty in the diagnosis of PCV. With the advent of combination therapy using both anti-VEGF therapy and photodynamic therapy for the treatment of PCV, the importance of distinguishing PCV from neovascular age-related macular degeneration—particularly occult choroidal neovascularization (CNV), with which it is often misdiagnosed—has increased. We therefore sought to evaluate the accuracy of SD OCT in detecting and differentiating PCV from occult CNV.
Patients and Methods
Inclusion Criteria and Data Collection
This observational case study was performed in accordance with the tenets of the Declaration of Helsinki and was approved by the Research and Ethics Committee of Moorfields Eye Hospital National Health Service Foundation Trust, London, United Kingdom.
We retrospectively reviewed 44 consecutive patients with 1 or more serous/hemorrhagic pigment epithelial detachments (PEDs) who visited the Medical Retina Service at Moorfields Eye Hospital between January 2012 and December 2012. Subjects included in this study were required to have 1 or more PEDs in at least 1 eye, with SD OCT and ICGA performed at the same visit. We excluded from the study patients with the following diagnosis: classic exudative age-related macular degeneration (which often presents without PEDs and has clear fluorescein angiographic features), myopic CNV and other secondary CNVs, and central serous chorioretinopathy (CSCR).
All patients with serous/hemorrhagic PED underwent slit-lamp examination with noncontact fundus lens, color fundus photographs, SD OCT, fundus fluorescein angiography (FFA), and ICGA in both eyes. Color fundus photographs, FFA, and ICGA were performed with a Topcon TRC 50 DX Fundus camera and Spectralis HRA+OCT (Heidelberg Engineering GmbH, Heidelberg, Germany). Spectral-domain optical coherence tomography was conducted using the Spectralis HRA+OCT (Heidelberg Engineering GmbH, Heidelberg, Germany) with viewing module version 184.108.40.206 (Heidelberg Engineering GmbH, Heidelberg, Germany). The SD OCT protocol included a dense horizontal linear scan centered on the fovea and some patients had peripapillary scans as well.
Definition of Polypoidal Choroidal Vasculopathy Through Qualitative Analysis of Optical Coherence Tomography Images
All SD OCT images were consecutively evaluated by 1 of the authors (G.D.S.) masked to the results of patient characteristics, history, ICGA imaging, diagnosis, and treatment. The author diagnosed PCV vs occult CNV based on the presence of at least 3 of the following findings: multiple PEDs, sharp PED peak, PED notch, and rounded hyporeflective area representing the polyp lumen within the hyperreflective lesions adherent to the underside of the retinal pigment epithelium (RPE). The presence of hyperreflective intraretinal hard exudates was also documented. Occult CNV was defined according to Macular Photocoagulation Study criteria as fluorescein angiographic evidence of a fibrovascular PED (stippled hyperfluorescence) or late leakage of undetermined source.
ICGA and FFA images were reviewed by 2 authors (G.D.S., G.L.) masked to the results of the SD OCT grading as well as all patient characteristics to determine the presence of polyps or occult CNV. Disagreements were resolved by open adjudication between the 2 authors.
Statistical analysis was performed using Statistical Analysis System (SAS, Statistical Analysis Software, New Jersey, USA) software. The Mann-Whitney U nonparametric test was used to compare age between the 2 groups (PCV vs occult CNV). Fisher exact test was used to compare SD OCT characteristics that differed between these 2 groups. P < .05 was considered statistically significant.
Baseline Characteristics of Patients and Eyes Analyzed
In this study, we reviewed 88 eyes from 44 consecutive patients with 1 or more serous/hemorrhagic PEDs who visited the Medical Retina Service at Moorfields Eye Hospital between January 2012 and December 2012. Three eyes were excluded because of disciform scars and poor-quality images attributable to poor fixation and 34 were within normal limits. Among the 51 eyes included in the study, 7 presented with bilateral disease (5 PCV in both eyes, and 2 occult CNV in both eyes). All eyes had good-quality SD OCT and ICGA.
The patients consisted of 14 men and 30 women aged between 48 and 95 years (median 70 years). The qualitative tomographic SD OCT review showed signs of PCV in 35 eyes of 32 patients (11 men and 21 women), with 5 patients presenting with bilateral disease. PCV prevalence was higher in female subjects, with 65.6%, vs 34.4% male. All the polypoidal lesions were localized within the PEDs. A diagnosis of PCV was confirmed using ICGA in 37 eyes ( Table 1 ). The median age of these patients was 69 years (range 48–91).
|Indocyanine Green Angiography|
|Spectral-Domain Optical Coherence Tomography|
Of 14 remaining eyes graded by 1 of the authors, SD OCT tomographic findings of occult CNV were detected in all of them, but only 13 were confirmed to have occult CNV on ICGA, whereas 1 had a polypoidal lesion (ie, 1 false positive) ( Table 1 ). The median age of this second group was 79 years (range 59–95), which was significantly higher than that of patients with PCV ( P = .02).
The specificity for the detection of PCV using SD OCT was 92.9% ( Table 2 ), with sensitivity of 94.6% ( Table 2 ) and positive and negative predictive values of 97.2% and 86.7%, respectively.
|Imaging Modality||Sensitivity (%)||Specificity (%)||Positive Predictive Value (%)||Negative Predictive Value (%)|
|Spectral-domain optical coherence tomography||94.6||92.9||97.2||86.7|
Assessment of Spectral-Domain Optical Coherence Tomography Features of Polypoidal Choroidal Vasculopathy
In the current study, of the 37 eyes with a confirmed diagnosis of PCV at ICGA, at least 3 of the following SD OCT tomographic diagnostic criteria were required to make a definite diagnosis of PCV: a sharp PED peak ( Figures 1 [Middle], 2 [Top middle], and 3 [Top middle] ), present in 34 eyes (91.9%); a PED notch ( Figures 1 [Bottom], 2 [Top middle], 3 [Bottom middle], 4 [Bottom], and 5 [Middle] ), in 37 eyes (100%); a visible hyporeflective lumen within hyperreflective lesions adherent to the outer surface of the retinal pigment epithelium ( Figures 1 [Middle], 2 [Bottom], 3 [Top middle], 4 [Middle and Bottom], and 5 [Middle and Bottom]), in 35 eyes (94.6%); or multiple PEDs, seen in 35 eyes (94.6%). Among the 14 eyes diagnosed with occult PED, only 1 (7.1%) presented a sharp PED peak, while none had a visible hyporeflective lumen or a PED notch and 4 (28.6%) presented with more than 1 PED. Hard exudates ( Figures 1 [Middle], 2 [Bottom middle], 3 [Bottom], 4 [Middle], and 5 [Middle]) were noted in 32 eyes (86.5%) with PCV, while only 6 eyes (42.9%) with occult CNV had these lesions ( Table 3 ). All these findings were statistically significantly different between PCV and occult CNV ( P < .0001). Furthermore, SD OCT of occult CNVs showed shallow PEDs and areas of medium reflectivity organized in horizontal bands underneath the RPE ( Figure 6 ), with no evidence of hyporeflective lumen or tomographic suspicion of PCV. The multimodal study of these occult CNVs with FFA and ICGA confirmed irregular late leakage suggestive of fibrovascular PEDs ( Figure 6 , Top left, Top right, and Top middle).