The Role Of OCT-Based Biomarkers For Predicting Treatment Response In Eyes With Centre Involving Diabetic Macular Oedema Treated With Anti-VEGF Injections

AIM

To investigate the associations between SD-OCT biomarkers and the treatment response among DMO eyes

treated with intravitreal anti-VEGF injection

OBJECTIVES

1. To assess the correlation with baseline SD-OCT biomarkers and baseline VA

2. To assess correlation between baseline SD-OCT biomarkers and subsequent VA after anti-VEGF

treatment

3. To compare SD-OCT biomarkers in DMO patients before and after anti-VEGF injection

MATERIALS AND METHODS

STUDY DESIGN

Longitudinal prospective cohort study

STUDY POPULATION:

Inclusion criteria:

1. Patients with either type 1 or type 2 DM aged 18 years or older, who have consented for the study

2. CI-DMO, defined by SD-OCT foveal involvement of abnormal intraretinal and/or subretinal fluid

with concurrent thickening affecting the 1mm diameter of CST

Exclusion criteria:

1. Pre-existing retinal or macular disease other than DR or DMO

2. SD-OCT images with poor quality that was insufficient for assessment

3. Pre-existing ocular diseases such as cataract, glaucoma, pathological myopia and uveitis

4. Any cataract surgery done in past 3 months

5. Prior vitreoretinal surgery or intravitreal anti-VEGF injections

6. Patients who have undergone any supplementary macular laser and/or intravitreal steroid implants

STUDY DURATION

18 months (one and a half years)

STUDY SETTING:

Retina clinic at Nethradhama Superspeciality Eye Hospital, Jayanagar, Bengaluru- 560070. The hospital

is equipped with Spectral domain OPTICAL COHERENCE TOMOGRAPHY, "HEIDELBERG HRAI

SPECTRALIS"

SAMPLE SIZE

Sample size calculation was based on the principal of “Inference for a mean, comparing a mean to know a

value.’ To calculate the sample size, a reference study was used. The mean baseline best corrected visual

acuity was 0.62 Log MAR units and after the last anti-VEGF injection was 0.49 Log MAR units, respectively.

The standard deviation was 0.3. Considering 80% power (alpha = 0.05), and a precision error of 5% to detect

difference of 20% or more in BCVA, the estimated sample size was calculated to be 44.

PROCEDURE:

All treatment naïve patients with diabetes with CI-DMO, who have been indicated intravitreal anti-VEGF

injections will be recruited for this study. After explaining the nature and purpose of the study, an informed

consent will be taken. Their baseline demographics such as age, gender, duration of DM, HbA1c levels and

other associated co-morbidities will be taken. Baseline ophthalmic examination including VA measured with

Snellen’s chart and converted to Logarithm of the Minimum Angle of Resolution (logMAR) scale, severity

of DR and DMO status will be recorded. VA at baseline, `, 3 and 6 months will be recorded. The number

and type of anti-VEGF injections will also be noted. OCT will be done prior to anti-VEGF therapy and at 1

month, 3 months and 6 months follow-up. The OCT will be assessed based on the below mentioned

parameters.

OCT Image Analysis

All OCT images will be analysed by 2 retina specialists.

Quantitative measurement, namely the CST was provided by the in-built Spectralis software using the volume

scan, consisting of 25 lines horizontal raster scans covering 20°×20°, centred on fovea.

Qualitative analysis of the OCT images was graded based on the scan passing through the fovea centre and

the 3 B-scans immediately superior and inferior to the fovea. The following qualitative parameters will be

analysed

1. Morphology of DMO.

2. Vitreomacular relationship.

3. The presence of DRIL.

4. The presence and number of hyper-reflective retinal foci (HRF).

5. The presence of Hyper-reflectivity within the intra-retinal cyst.

6. The visibility and continuity of ELM and EZ.

Morphology of DMO was classified based on OCT findings into (1) CMO, defined as presence of intraretinal

cystoid spaces; (2) Diffuse retinal thickening (DRT), defined as diffuse sponge like thickening of retina and

(3) Serous retinal detachment (SRD), defined as the presence of subretinal fluid. Patients may have one or

more of these features concurrently.

Vitreomacular relationship will be classified according to the International Vitreomacular Traction Study

Group classification28 into (1) no visible vitreoretinal adhesion; (2) incomplete vitreous detachment; (3)

complete posterior vitreous detachment; (4) vitreomacular traction; and (5) epiretinal membrane.

DRIL is defined as the inability to segment the boundaries of the OPL, INL and the GCL-IPL complex on the

central 1 mm zone of horizontal scan centred on fovea. It was graded as either absent or present23.

Presence of hyper-reflectivity within the intra-retinal cyst is described as any hyper-reflectivity seen within

the cystic spaces. It’ll be indicated as present/absent.

Hyper-reflective foci (HRF) is defined as circumscribed dots with similar reflectivity in the retinal nerve fibre

layer, absence of back-shadowing and <30μm diameter. HRF is defined as hyper-reflective foci between

internal limiting membrane and RPE. HRF is not to be confused with hard exudates which arises from lipid

exudation. The number of hyper-reflective foci will be counted manually based on the 1mm zone of the

horizontal scan centred on fovea21. HRF will be graded into three groups: no hyper-reflective foci, 1–10 hyperreflective

foci and >10 hyper-reflective foci, respectively.

The visibility and continuity of ELM and EZ will be graded within the central 1mm wide area of the horizontal

B-scan centred on fovea centre. ELM and EZ are defined as the first three hyper-reflective bands of the four

outer- most layers on SD-OCT, with the third band being COST and the fourth hyper-reflective band being

RPE. Each layer will classified into intact, disrupted (partially visible and discernible) and absent (not

discernible at all). The ELM and EZ will be interpreted as a unit because in the presence of subretinal fluid

where EZ cannot be identified, the integrity of ELM would be graded instead21.

STATISTICAL ANALYSIS

Data will be entered in MS excel and analysed using SPSS software. Qualitative data will be presented in the

form of Proportions and pie diagrams, bar charts will be used to represent graphically. All categorical variables

will be summarised using frequency and percentage. Continuous variables will be summarised using mean

and standard deviation if data follows normality, else, median and IQR will be used. Student’s t test will be

the test of significance for quantitative data and chi-square test will be the test of significance for qualitative

data. p value <0.05 will be considered as statistically significant.