English

Abstract

BACKGROUND We wanted to measure and compare retinal sensitivity in central 30 degree in diabetic patients, with and without diabetic retinopathy in different stages, evaluate changes in retinal sensitivity in relation to change in HbA1c values, measure and compare GCL thickness in various stages of DR with the help of optical coherence tomography (OCT). METHODS This observational, cross-sectional study involving 100 eyes of 100 middle aged (45 - 64 years) type 2 diabetes mellitus patients (50 eyes without DR - group 1 & 50 eyes with DR - group 2) without any other ocular abnormalities was conducted in the outpatient and in-patient departments of department of ophthalmology in a tertiary care centre in West Bengal. Non-randomised sequential sampling was performed with corrected visual acuity better than or equal to 6 / 12. Fasting and postprandial blood glucose and HbA1c were estimated. Detailed ocular examination was performed using direct and indirect ophthalmoscope with + 20 D lens and slit lamp bio-microscope using + 90D lens. Retinal sensitivity was assessed by Humphrey visual field analyser by Swedish Interactive Thresholding Algorithm (SITA) standard strategy (30 - 2 programme). Spectral domain optical coherence tomography (SD OCT) was performed in all patients. SPSS version 20 has been used for the analysis. RESULTS Among DR patients, 33 had mild non-proliferative diabetic retinopathy (NPDR) (male = 15, female = 18) and 17 had moderate NPDR (male = 7, female = 10). Mean age in DR (Gr. 2) and no DR (Gr. 1) group were 52.62 and 50.74 years respectively. Mean foveal sensitivity and mean retinal sensitivity decreased significantly (P-value 0.001 and 0.002 respectively) in group 2 patients. It further decreased with increased severity of DR. Mean ganglion cell + inner plexiform layer (GC + IPL) thickness in temporal quadrant decreased in DR group compared to no DR group with significant difference between the two (P-value < 0.001). Mean retinal nerve fibre layer (RNFL) thickness was significantly reduced in DR group (P-value < 0.001). HbA1c mean in no DR (6.7 %) and DR group (8.07 %) and in mild (7.5 %) and moderate NPDR (9.17 %) shows significant association of poor control of blood sugar with severity of DR. CONCLUSIONS Retinal sensitivity decreased significantly in diabetes even without retinopathy as detected by automated perimetry. Significant decrease in retinal thickness as detected by OCT suggested that neurodegeneration occurs in diabetes even without retinopathy. So automated perimetry and OCT could be helpful in identifying persons at an early stage who are at risk of future vision loss due to diabetes. KEYWORDS Diabetic Retinopathy, Retinal Neurodegeneration, Retinal Sensitivity, Automated Perimetry, GCL + IPL Thickness, RNFL Thickness