Association between Optical Coherence Tomography and Fluorescein Angiography based retinal features in the diagnosis of Macular Edema
Graphical abstract
Introduction
The build-up of fluids in the macular region occurs when the damaged blood vessels inside the retina leak fluid and protein deposits in the macular region. This leads to tissue swelling and is called macular edema (ME). ME has multiple aetiologies such as diabetes, and in patients with diabetes, diabetic ME is a leading cause of central vision loss. Optical coherence tomography (OCT) has been in use for the last two decades, has allowed higher resolution imaging of the retina and is now routinely used for the detection of ME [[1], [2], [3]]. OCT allows the objective measurement of retinal thickness and localises the fluid within the retina. The analysis of ME is aided by fluorescein angiography (FA) that identifies the site, type and amount of leakage using fluorescein dye, whereas OCT measures the retinal thickness and localises it within the retinal layers [4].
ME requiring treatment (MERT) eyes are characterised by retinal thickening, presence of hard exudates, presence of leakage patterns on FA images and structural changes on OCT images. Traditionally, patients undergo a biomicroscopic examination using a slit lamp as well as indirect ophthalmoscopy and are judged with stereo biomicroscopy, but these methods lack high resolution imaging. Owing to the improved resolution and speed of OCT and FA, most ophthalmologists currently use these for this purpose. When using OCT, ME is considered requiring treatment based on OCT features such as the thickness of the macula, changes in the morphology of the retina and macular traction [5]. Other guidelines such as the presence of retinal thickening and hard exudate at or within 500 μm of the centre of the macula are also used to define MERT [6]. Previous studies have suggested that the thickness of the retina as measured by OCT and retinal volumes within radii of 0.5 and 1.11 mm of fixation are the best objective measures to identify patients requiring treatment [7]. However, there is lack of consensus in the criterion for MERT.
This study has investigated the association of the measurements obtained from OCT and FA in the diagnosis of ME (MERT and non-MERT). The scans were graded as MERT and non-MERT using the accepted interpretation criteria by two clinical experts from two different hospitals in Melbourne who were independent of each other [8]. The study aimed to determine the association between the objective OCT and retinal vessel geometrical parameters obtained from FA and the ME classification to determine the most relevant features to categorise the eyes as MERT and non-MERT without any bias.
Section snippets
Materials
This study investigated the OCT and FA data of treatment-naive patients who visited Essendon Eye Clinic, Victoria, Australia with suspected ME irrespective of the aetiologies: diabetes, central or branch vein occlusion, uveitis and dye leakage associated with choroidal neovascularisation syndromes. The study was approved by the RMIT University Human Experiments Ethics Committee and was conducted following the Helsinki accord 1986 (modified 2004).
FA was performed using the Optos 200TX
Methods
The study investigated the retinal vasculatures surrounding the macular region. The selection of the macular region on the FA was based on the ETDRS [12], which is a commonly accepted grid for such recordings and which considers the fovea as the reference. Instead of using traditional methods based on slit lamp or stereo fundoscopy, OCT-based analysis was used. For each FA image, the fovea centralis position was manually marked by the ophthalmologist as the point with the lowest pixel intensity
Statistical analysis
Statistical analyses were performed using MedCalc 10.0.2.0 (MedCalc Software Ostend, Belgium) for the nine parameters, four obtained from OCT and the five from the selected FA retina image frame. Statistical distribution for each feature was obtained and evaluated using the Shapiro–Wilk test. Kruskal Wallis test was performed for the data that did not satisfy the normality test to determine the statistically significant group difference between MERT and non-MERT. In all analyses, p < 0.05 was
Results
Overall, 52 eyes were classified as MERT and 29 eyes as non-MERT by clinicians based on the inspection of the FA and OCT B-scans. The results of the statistical analyses of the parameters obtained from OCT and FA images are presented in Table 1. This table shows the median, standard deviation, average rank and p value using the Kruskal Wallis test for all the nine parameters. The results showed that there was a significant difference between the two groups for seven of the nine parameters. ABA
Discussions
Retinal vasculature characteristics, i.e. FD and tortuosity, have been associated with cardiovascular diseases, stroke and diabetes. ME, a major cause of vision impairment, appears owing to the disruption of the blood–retina barrier, which damages the retinal vasculature and leads to the accumulation of fluid beneath the surface of the retina, thus causing retinal thickening.
OCT captures a high-resolution 3D cross-sectional image of the retina, making it suitable for detecting retinal diseases
Conclusion
We have found that retinal thickness obtained from the OCT image is substantially correlated with the changes in the retinal vasculature on the FA image. This offers an alternate diagnosis and a potential predicator to the changes in retinal thickness resulting from ME and an alternative for the detection of ME using FA images. Moreover, it was observed that OCT parameters such as central retinal thickness and average para-fovea thickness are suitable for detecting MERT. These results reveal
Declaration of competing interest
All authors declare that they have no conflict of interest.
Acknowledgement
The authors would like to thank Olivija Tsaketas and the entire staff of Essendon Eye Clinic, Melbourne, Australia, for their valuable time, advice and support during data collection and the entire study.
References (23)
- et al.
Fluorescein angiography versus optical coherence tomography for diagnosis of uveitic macular edema
Ophthalmology
(2013) Non-linear dynamics for clinicians: chaos theory, fractals, and complexity at the bedside
Lancet
(1996)- et al.
Computer-automated quantification of plus disease in retinopathy of prematurity
J. AAPOS
(2003) - et al.
Optical coherence tomography to detect and manage retinal disease and glaucoma
Am. J. Ophthalmol.
(2004) - et al.
Diabetic macular edema
- et al.
Guidelines for the management of diabetic macular edema by the European society of retina specialists (EURETINA)
Ophthalmologica
(2017) Optical coherence tomography (OCT) in macular edema
Doc. Ophthalmol.
(1999)- et al.
Optical coherence tomography for macular edema. Classification, quantitative assessment, and rational usage in the clinical practice
Ophthalmologe
(2004) - et al.
Types of diabetic macular edema assessed by optical coherence tomography
Folia Med.
(2008) Photocoagulation for diabetic macular edema: early treatment diabetic retinopathy study report number 1
Arch. Ophthalmol.
(1985)
Optimal optical coherence tomography–based measures in the diagnosis of clinically significant macular edema: retinal volume vs foveal thickness
Arch. Ophthalmol.
Cited by (6)
Spatially adaptive blind deconvolution methods for optical coherence tomography
2022, Computers in Biology and MedicineCitation Excerpt :OCT uses low coherence light to collect images with a penetration depth of a few millimeters and a spatial resolution of tens of micrometers in vivo tissue. Compared with traditional imaging techniques, e.g., X-Ray computerized tomography (X-CT) [5], magnetic resonance imaging (MRI) [6], and ultrasound imaging (UI) [7], OCT has the advantages of being a noncontact and nondamaging method with high spatial resolution and low cost [8] and, thus, has been widely used in medical areas such as ophthalmic disease [9,10], cardio-cerebrovascular disease [11], dermatology [12] and gastroenterology [13]. Resolution enhancement is one of the most concerning issues in OCT.
Deep learning based joint segmentation and characterization of multi-class retinal fluid lesions on OCT scans for clinical use in anti-VEGF therapy
2021, Computers in Biology and MedicineCitation Excerpt :Since the last two decades, the field of retinal image processing has received much attention, with an emphasis on OCT-based automated retinal analysis. Researchers have suggested several automated algorithms for the following main tasks relevant to retinal OCT imaging: (1) elimination of speckles and OCT denoising [17,18], (2) chorioretinal layers segmentation [19–21], (3) segmentation of anatomical lesions [21–25], and (4) classification of retinal pathologies [25–30]. The segmentation of MRF lesions in OCT imaging is realized as a difficult task for two main reasons: 1) high signal-to-noise ratio (SNR) variability between the scans of different OCT vendors (Topcon, Cirrus, and Spectralis); and 2) the considerable discrepancies in volume, structure, and delineation of the MRF lesions [13].
A review of methods for automatic detection of macular edema
2021, Biomedical Signal Processing and ControlCitation Excerpt :Spectral OCT images were used and good results are reported. Ajaz et al (2020) [124] performed fractal analysis of the FA of retinal vessels in the macular region and found that this was associated with the changes in retinal thickness. Their work discovered that OCT parameters can potentially be used for directly identifying ME.
Weakly supervised detection of central serous chorioretinopathy based on local binary patterns and discrete wavelet transform
2020, Computers in Biology and MedicineCitation Excerpt :At present, there are few researches on the detection of CSCR. Most researchers have layered and segmented the OCT B-scan images [8–11] which are used for the diagnosis of common fundus diseases such as age-related macular degeneration (AMD), diabetic macular edema (DME), diabetic retinopathy (DR), glaucoma and so on, as well as location of the associated lesion areas. Some scholars have segmented the CSCR lesions in OCT B-scan images of fundus and achieved good segmentation results.
ISnoDi-MDRF: Identifying snoRNA-Disease Associations Based on Multiple Biological Data by Ranking Framework
2023, IEEE/ACM Transactions on Computational Biology and Bioinformatics