Abstract
Purpose: :
Since implementation of the English National Screening Programme for Diabetic Retinopathy (ENSPDR) 2006, number of referrals to hospital eye services has increased markedly, 60-70% estimated to be for maculopathy. In some programmes, optical coherence topography (OCT) assessment of macular thickening is used to reduce work-load by diverting non-thickened maculopathy referrals away from hospital eye clinics. Our aim was to assess efficacy of time-domain OCT combined with digital macular images in a ‘virtual’ examination to distinguish macular thickening from maculopathy without thickening and to look at follow up outcomes of these patients.
Methods: :
In 2010, 100 consecutive maculopathy referrals due to ‘exudates’ within 1dd of the fovea were identified (R3 retinopathy and M1 vision referrals excluded). Retrospective examination of Zeiss Stratus OCT images and clinic Topcon digital macular photographs resulted in a virtual diagnosis for each referred macula: M0=no exudate, no thickening; M1=exudate, no thickening; CSMO (clinically significant macular oedema)=exudate plus thickening. Specifically, OCT slices through exudates were examined for oedema, and macular thickness map for thickening. Virtual diagnosis was compared to clinic examination diagnosis. Those with CSMO on clinical examination underwent macular laser treatment. Outcomes for these patients at 6 months were examined.
Results: :
119 eyes of 100 patients were given virtual (v) and examination (e) diagnoses. 27(24%) were M0v/M0e; 2(2%) M1v/M0e; 19(16%) M1v/M1e; 31(26%) M1v/CSMOe; and 40(34%) CSMOv/CSMOe. OCT therefore only identified 40 of 71 eyes (56%) with clinical CSMO requiring macular laser. Central macular thickness was significantly different at baseline between CSMOv/CSMOe and M1v/CSMOe groups (p=0.001). At 6 months after laser in the M1v/CSMOe group, 21% had CSMO on examination and OCT, 14% had CSMO on examination but not OCT, 29% had exudate but no thickening on either examination or OCT, and 36% had no exudate or thickening. Therefore 35% required more laser, compared to 68% in the CSMOv/CSMOe and 10% in the M1v/M1e group. There was no group of patients with M1v/CSMOe who did not undergo laser for comparison of 6 month outcome, but the difference in progression between the M1v/CSMOe and M1v/M1e groups supports the assumption that the thickening seen on clinical examination was real and required laser treatment.
Conclusions: :
When combined with macular photography, Stratus OCT is a useful predictor of maculopathy requiring laser but in this study it could not be substituted for clinical examination.
Keywords: diabetic retinopathy • macula/fovea • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)