20th Meeting of the European Association for the Study of Diabetes Eye Complications Study Group (EASDec)

Paris, France  21st -22nd May 2010


R. Muniraju1,2, J. Ramu2, R. Duggan2
1Mayday University Hospital, Croydon - United Kingdom
2Diabetic Retinopathy Screening Service, Croydon – United Kingdom

INTRODUCTION. National Screening Programme for Diabetic Retinopathy has set 20 quality standards, the first of which is to reduce new blindness due to Diabetic Retinopathy (DR) by a minimum of 10% within 5 years.
PURPOSE. The national standard is Severely sight impaired (SSI) / Sight impaired (SI), compared to 1990/91 rates of 9.5 and 9.3 respectively per million per annum.To examine the incidence of SSI/SI from April 2008 to March 2009 due to DR in Croydon.
METHODS. A retrospective analysis of all 144 Certificates of Visual Impairment from Croydon during April 2008 to march 2009.
RESULTS. Total no of registrations due to DR=15, Severely sight impaired=8, Sight impaired=7, Croydon population =39,500, SSI per million per annum 23.6, SI per million per annum 20.6. The SSI and SI rate is above the national standard. The study conducted by Bunce C, during April 1999 to March 2000 has shown that the registrations of severe visual impairment due to DR was 38.4 per million per year.
CONCLUSIONS. The prevalence rate of diabetics in England is 3.9% as compared to the rate of 4.7% in Croydon. This could be because 41% of the Croydon population are Afro-Caribbean and other Ethnic Minorities as compared to 15% in England. Diabetes is 6 times more common in South Asians and up to 3 times more common among Afro-Carribeans. This probably explains the high registration rate of visual impairment due to diabetic retinopathy in Croydon.

K-J. Hellgren1,2, B. Bengtsson1, E. Agardh1
1Institution of clinical sciences, Ophthalmology, Skåne University Hospital - Sweden
2Clinic of Ophthalmology, County Hospital of Värmland - Sweden

INTRODUCTION. We have previously demonstrated that visual field correlates better to severity of diabetic retinopathy than visual acuity. To that end, we have created a new model for assessment of subtle visual field change in diabetes, analogous to that applied for glaucoma.
PURPOSE. To present baseline data of a study aiming at detecting subtle visual field change in diabetes.
METHODS. Eighty diabetic subjects with various degrees of non-proliferative retinopathy were consecutively recruited for examination during three to five years with retinal photography, fluorescein angiography, standard automated perimetry (SAP) 24-2 and central short wavelength automated perimetry (SWAP) 10-2.
RESULTS. Age at onset of diabetes was 44±15 years (mean±SD) and duration of diabetes 14±12 years. Forty-one subjects did not have any retinopathy, 17 had mild, 20 moderate, one severe and one proliferative retinopathy. SAP mean deviation, a global visual field measure in which 0dB represent an age- corrected normal value and negative values worse, was -0.39 dB for those without and -1,10 dB for those with retinopathy.
CONCLUSIONS. Our method is now being used in a prospective longitudinal study for early detection of subtle visual field change enabling functionally measurable effects of new treatment strategies addressing early stages of diabetic retinopathy.

R.L. Leigh
Hartlands Hospital, Selly Oak Hospital, Walsall Manor hospital, Walsall – United Kingdom

INTRODUCTION. Three Centres within the Birmingham and Black Country screening scheme have been assessing the efficacy and value of the NICE guidelines 2008 for Diabetic Retinopathy Screening in pregnancy.
PURPOSE. To assess whether digital diabetic screening in pregnant patients is safe and effective.
METHODS. In 33 months we obtained the screening results of 187 pregnant patients with diabetes and analysed the outcomes of their screening.
RESULTS. Of the total screened: 18 (10%) patients were referred to ophthalmology at their 1st screen. 94 (50%) developed no retinopathy throughout their pregnancy. 34 (18%) presented with background DR at their 1st screen, 25 patients remained stable throughout pregnancy, and 9 improved from background to normal in at least one eye. 18 (10%) deteriorated throughout their pregnancy, only 2 were referred to ophthalmology. 23 (12%) patients attended their 1st screen and did not attend any follow-up appointments, 8 of these patients had background DR at initial screen. A total of 20 (11%) patients were referred to ophthalmology during their pregnancy.
CONCLUSIONS. Digital screening for diabetic retinopathy throughout pregnancy (following NICE guidance) has proved clinically safe, with no adverse events and only 11% requiring referral to ophthalmology. This prevents most patients having to attend hospital eye clinic appointments, hence cost savings. One concern however was high non-attendance rates for which failsafe procedures need to be effective.

R. Hampshire, A.D. Wright, P. M. Dodson
Departments of Ophthalmology and Diabetic Retinal Screening, Birmingham Heartlands Hospital, Birmingham – United Kingdom

INTRODUCTION. Asteroid hyalosis (AH) is a degenerative condition with calcium soaps accumulating in spherical, discoid, creamy white or shiny bodies or strands in an otherwise normal vitreous. Disturbance of vision rarely occurs but retinal detail is often obscured. AH is a common cause of ungradable photographs within diabetic retinopathy screening programmes. The opacities are reported to move with the wavelike undulations of the vitreous but do not settle inferiorly.
PURPOSE. Assessment of any change in established asteroid hyalosis over time.
METHODS. 18 patients, 12 male 6 female, mean age 69.9 years (range 47-86), were studied who had three to five sets of photographs taken at annual intervals within the Birmingham and Black Country diabetic retinopathy screening programme.
RESULTS. Centration of the photographs was equivalent on each occasion. AH affected one eye only.
Visual acuity was 6/6-6/9 in 16 patients and 6/12 in two patients.
The number and extent of AH showed no change in 10 and increased in 8 eyes.
The pattern of AH was unchanged in 7, shifted in 6, and changed in distribution in 5 eyes
CONCLUSIONS. The extent of AH increased over a three year period of observation in nearly half the patients and the pattern of AH also changed in a significant proportion, particularly in those with a few lesions. AH continued to develop in the majority of patients and although some movement of the bodies occurred, this was not sufficient to permit retinal grading.

M. Trento1, L. Charrier2, M. Montanaro3, F. Bandello4, L. Scoccianti5, F. Cavallo6, M. Porta7
1Laboratory of Clinical Pedagogy, Department Internal Medicine, University of Turin – Italy
2Department of Public Health and Microbiology, University of Turin – Italy
3Diabetic Retinopathy Centre, Department Internal Medicine, University of Turin – Italy
4Department of Ophthalmology, Hospital San Raffaele, Milano – Italy
5Diabetic Retinopathy Centre and Laser Treatment, Institute of Ophthalmology, Parma – Italy
6Department of Public Health and Microbiology, University of Turin – Italy
7Laboratory of Clinical Pedagogy, Diabetic Retinopathy Centre,University of Turin - Italy

INTRODUCTION. Retinopathy is a complication of diabetes that requires regular control visits and, sometimes, treatment by laser photocoagulation.
PURPOSE. To evaluate changes in vision-related quality-of-life in patients with diabetic retinopathy (DR) using the 25-item National Eye Institute Visual Functioning Questionnaire (NEI VFQ-25) in 3 Centres for DR screening.
METHODS. VFQ-25 was self-administered between 2006-2009 to 146 patients (Pz) with visual acuity (va)<5/10 in the better eye: 43 in Turin, 52 in Parma, and 51 in Udine. VFQ-25 items explored 12 scales: General Health (GH) and Vision (GV), Ocular Pain (OP), Near Activities (NA), Distance Activities (DA), Vis Specific Social Functioning (VSSF), VisSpecMentalHealth (VSMH), VisSpecRole Difficulties (VSRD), VisSpec Dependency (VSD), Driving (D), Color Vision (CV), Peripheral Vision (PV). Data on socio-anagraphic variables and presence of DR, cataract and laser treatment (LT) were collected.
RESULTS. Pz in the 3 Centres did not differ by age, gender and diabetes duration. Pz in Parma had not had cataract interventions and had more maculopathy (p<0.0001). Va was associated to the GV, NA, DA, VSSF, VSRD, VSD, D, CV (p<0.01), VSMH (p<0.05) and PV (p=0.01) scores. Laser treatment (-12.5; p=0.002) and centre (Parma-vs-Torino: -15.14; p=0.021; Udine-vs-Torino: -10.62; p=0.038) were relevant to GH. LT was associated to lower scores in GV (-9.32; p=0.004) and Driving (-15.6; p=0.021). Afference to centre was associated to significant variations in the GV, OP, DA scores, with lower scores in Parma and Udine than Turin. Women obtained significantly better scores than men in CV (+15.91; p=0.005).
CONCLUSIONS. Ocular complications and loss of vision modify the way people perceive their own ability to function autonomously.


A.G. Meyramova, F.A. Mindubaeva, G.G. Meyramov, R.M. Meyramova

Karaganda State medical University - Kazakhstan

INTRODUCTION. Nineteen potential Diabetogenic Zinc-binding Cytotoxins (DZC) were isolated from B-cell Toxic Complexes (BTC) that can destruct B-cells within 15-20 min. Eighteen are derivatives of 8-oxyquinolin and one is of dithizone (DZ). Retinal zinc and DZC are potential risk factors in diabetic retinopathy (DR).

PURPOSE. Our aim was to prove that DZCs can be formed and affect the retina which can be attenuated by the removal of zinc.

METHODS. Rabbits were treated with intra venous injection of DZ [49.3-52.6mg/kg] or 8-para(toluenesulphonylamino)quinolin (8PTSQ) [47.6-48.9mg/kg] to induce toxicity. Protection was assessed by IV co-application of sodium diethyldithiocarbamate (DDCA) [492.5-505.2mg/kg] with 8PTSQ [48.9-50.2mg/kg], DDCA [489.7-501.9mg/kg] with DZ [46.3-49.5 mg/kg] or cystein [902 mg/kg] with DZ [47.7 mg/ kg]. Retina and pancreatic islets of 6 rabbits in each group were examined histologically using aldehydefuschine, insulin immunohistochemistry, preudoisocyanine and DZ staining. Results were compared to untreated controls.

RESULTS. DZ and 8PTSQ treatment alone resulted in: 1) the destruction of islets, 2) decreased tissue insulin content (1.12±0.04 vs 1.89±0.06) and 3) increased the granulated DZ staining both in B-cells and in retina. Co-application of DDAC largely prevented the DZ or 8PTSQ induced changes, though a slightly decrease in insulin content could be observed (1.79±0.06 vs 1.93±0.04). Cystein co-application had a less marked effect.

CONCLUSIONS. Removal of zinc with DDCA or cystein protect retina from DZC toxicity suggesting that limiting DZCs access to zinc might be important to harness or reduce the severity of DR.

K.K. Tilma, T. Bek
Department of Ophthalmology, Århus University Hospital, Århus - Denmark

INTRODUCTION. Changes in the retinal arteriolar diameter and retinal autoregulation are involved in the pathophysiology of diabetic retinopathy. In vitro studies have shown that prostaglandins are involved in the tone regulation of retinal arterioles.
PURPOSE. To study whether prostaglandin mediators can affect the diameter of retinal arterioles in diabetic patients.
METHODS. In a double-blinded randomised cross over study 24 normal persons and 22 age and sex matched diabetic patients with minimal diabetic retinopathy were randomised to receive treatment with either topical Xalatan for one week followed by a wash out period and treatment with topical Voltaren for another week, or to receive the two drugs in the reverse order. The baseline diameter of retinal arterioles, the diameter response to an increase in the arterial blood pressure and to stimulation with flickering light, or to both, were studied using the Retinal Vessel Analyzer (RVA).
RESULTS. In diabetic patients the baseline diameter of retinal arterioles was 134,0±15,8 microns and was significantly reduced by both Xalatan (to 130,0±15,1 microns, p=0,009) and by Voltaren (to 132,5±15,6 microns, p=0,03). In normal persons the baseline diameter of retinal arterioles was 118,6±6,0 microns which was not significantly changed by the two treatment regimes (p=0,4 and p=0,1 respectively). Neither Xalatan nor Voltaren induced any changes in the autoregulatory responses.
CONCLUSIONS. Topical treatment with Xalatan and Voltaren for one week contracts retinal arterioles in type 1 diabetic patients with minimal retinopathy. These compounds may be possible future treatments for retinal hyperperfusion in diabetic retinopathy.

J. Figueira1,2, S. Leal2, A. Rosa2, J. Mira2, R. Silva1,2, J. Murta2
1AIBILI - Association for Innovation and Biomedical Research on Light and Image, Coimbra - Portugal 2Department of Ophthalmology, University Hospital of Coimbra - Portugal

INTRODUCTION. Cataract surgery is associated with worsening of diabetic macular oedema (DME).
PURPOSE. The authors evaluate the efficacy and safety of intraoperatory administration of intravitreal triamcinolone (IVT) on cataract surgery of diabetic patients with DME.
METHODS. Prospective, institutional, randomised study (1:1) of diabetic patients needing cataract surgery. Group A was administered IVT (4mg/0,1ml) at the end of cataract surgery. Group B was subject to cataract surgery alone. All patients were observed preoperatively (PO), at the 1st day (D1), and after 1, 6 and 12 weeks (W1, W6 and W12), undergoing at each visit ophthalmological examination and OCT (Stratus OCT).
RESULTS. On group A (n=19) average VA was 1.17 logMAR (PO), 0.53 logMAR (W1), 0.43 (W6) and 0.40 logMAR (W12), and mean central OCT thickness was 352µm (PO), 313µm (W1), 288µm (W6) and 300µm (W12). On group B (n=19) average VA was 0.70 logMAR (PO), 0.40 logMAR (W1), 0.37 logMAR (W6) and 0.40 logMAR (W12), and central OCT thickness was 342µm (PO), 349µm (W1), 417µm (W6) and 413µm (W12). At 12 weeks, on group A 90% of patients improved ≥0.3 logMAR and on group B 53% of patients improved ≥0.3 logMAR.
CONCLUSIONS. In our study IVT was an effective measure to control DME in diabetic patients submitted to cataract surgery, increasing from 52% to 90% the percentage of patients gaining 0.3 logMar or more. The central OCT thickness had a significant decrease on the Phaco+IVT group compared with the cataract surgery alone.

S. Feldman-Billard1, B. Dupas2, N. Sedira1, T. Meas3, A. Erginay2, P.J. Guillausseau3, P. Massin2
1Service de Médecine Interne, CHNO des Quinze-Vingts, Paris – France
2Service d'Ophtalmologie, Hôpital Lariboisière, Paris - France.
3Service de Médecine Interne, Hôpital Lariboisière, Paris - France

INTRODUCTION. Although spontaneous diurnal variations in diabetic macular oedema (DME) measured by Optical Coherence Tomography (OCT) have been previously reported, their causes remain unknown.
PURPOSE. To evaluate spontaneous diurnal variation of central macular thickness (CMT) and its relation to glucose and blood pressure changes in patients with DME.
METHODS. Prospective observational study in 22 diabetic patients (mean age: 60±12yrs) with DME involving the centre of the macula. OCT (Carl Zeiss, Meditec, Inc, Dublin, CA) scans was performed at 9am, 12am, 3pm, 6pm and the day after at 9am. During the same day, protocol included concomitant ambulatory 24h-blood pressure (NOVACOR Diasys Integra II, France) and 24h-glucose monitoring (CGMS, Medtronic, Minneapolis, MN).
RESULTS. CMT decreased consistently from 9am to 6pm in ten patients (374 to 337 µm) and increased or remained steady in twelve (383 to 390 µm), with a significant difference in CMT absolute change between the two groups (p<0.001). The lower the mean diurnal blood glucose (CGMS data), the smaller the decrease in CMT during the day (p=0.027). Moreover, 8/12 patients (67%) with no CMT decrease experimented a diurnal hypoglycaemia (<60 mg/dl) reported from CGMS data vs none with CMT decrease (p=0.002). Furthermore, hypoglycaemic episodes were associated with a median 27% increase (from 119±19 to 151±26 mmHg) of mean systolic blood pressure.
CONCLUSIONS. Hypoglycaemic events and their secondary blood pressure increases could explain the lack of diurnal CMT decrease in diabetic patients with DME.

M. Mehlsen, T. Bek
Department of Ophthalmology, Århus University Hospital, Århus - Denmark

INTRODUCTION. The presence of hard exudates increases the risk of developing vision threatening diabetic maculopathy. In several studies hard exudate formation has been shown to correlate with the level of serum lipids, but it is unknown whether these lipids constitute an independent risk factor for the development of vision threatening maculopathy.
PURPOSE. To study the association between serum triglyceride and the risk of reaching a treatment end point for diabetic retinopathy.
METHODS. 6079 examinations from 2452 patients (604 with T1D and 1848 with T2D) performed in the screening clinic at the Department of Ophthalmology, Århus University Hospital, and where results from the measurement of serum lipids (triglyceride, LDL, HDL and cholesterol) were subjected to weighed logistic regression to evaluate the contribution of these parameters to reaching a vision threatening treatment end point.
RESULTS. Serum triglyceride level was found to be an independent risk factor for reaching a treatment end point in T1DM (p=0.02), but not in T2DM (p=0.06), whereas no such relation was found for any of the other serum lipids examined.
CONCLUSIONS. Serum triglyceride should be included as an independent risk factor for reaching a treatment end point in T1D. This parameter can be used to optimise recommendations about control intervals in diabetic retinopathy screening.

M. Gupta, G. Lascaratos, A. Syrogiannis, A. Laude, B. Dhillon
Princess Alexandra Eye Pavilion Chalmers Street Edinburgh EH3 9HA – United Kingdom

INTRODUCTION. Rapid progression of diabetic retinopathy is associated with hypertension, pregnancy, poor control and duration of diabetes. Accelerated diabetic retinopathy in relation to HIV infection is uncommon.
PURPOSE. To report an unusual case of rapidly progressive diabetic retinopathy with severe visual loss and a poor response to laser therapy in a patient, who was later diagnosed to be Human Immunodeficiency Virus (HIV) positive.
METHODS. Observational case report.
RESULTS. This 32-year-old Caucasian female had been diagnosed with Type 2 diabetes recently and was referred to the eye clinic urgently following difficulties with her eyesight. On examination the visual acuity was 6/9 in the right and 6/12 in the left eye. Anterior segment examination was unremarkable. Dilated fundus examination revealed proliferative changes in both eyes. Fluorescein angiography showed diffuse staining with retinal ischaemia and leakage at the disc. Despite undergoing multiple sessions of bilateral pan retinal photocoagulation, she developed vitreous haemorrhage in both eyes and the visual acuity deteriorated to hand movements only. The progression from diagnosis of type 2 diabetes to vitreous haemorrhage in both eyes took less than 6 months. We also propose different pathophysiologic mechanisms to explain the rapid progression of proliferative diabetic retinopathy in a patient who is HIV positive.
CONCLUSIONS. This case not only describes the association between diabetes and AIDS, but also suggests the importance of HIV testing in a patient with unexplained rapidly progressive
diabetic retinopathy.

E.A. Wilkinson, H.J. Zambarakji
Eye Treatment Centre, Whipps Cross University Hospital Trust, London – United Kingdom

INTRODUCTION. What is the patient knowledge of recent HbA1c values in an urban diabetic retinal screening population and is it relevant?
PURPOSE. To highlight that, despite knowledge of recent HbA1c levels being a known determinant in patients' diabetic self-management and health outcomes, there is little known about levels of knowledge in a diabetic retinal screening population.
METHODS. Using a standardised questionnaire we conducted a prospective survey of HbA1c knowledge in 968 patients attending secondary care diabetic retinal screening clinics in a single urban eye unit. The UK NSC classification was used to grade the severity of the patients' retinopathy.
RESULTS. Of the 968 patients prospectively surveyed 909 (93.9%) were Type 2 diabetics and 69 (6.1%) Type 1 diabetics. Overall 97% of patients showed inaccurate or no knowledge of their HbA1c levels with greater awareness amongst the younger Type 1 diabetics. Type 2 insulin requiring diabetics showed no increased knowledge.
CONCLUSIONS. This study highlights the lack of knowledge of one of the key indicators of health outcomes in diabetes amongst a retinal screening population who are under multi-disciplinary care including both community fundus photography and hospital eye services. It suggests further work, including the correlation of knowledge of HbA1c levels and retinopathy grade and also the implementation of clear, concise and consistent advice given to patients as they pass from one discipline to the next.

M. Karthikeyan, H. Cilliers, L.P. Chow, D. Verma
Addenbrookes Hospital NHS Trust, Cambridge – United Kingdom

INTRODUCTION. Diabetic retinopathy is the third major cause of visual impairment in UK. Screening services attempt to reduce the incidence of severe blindness by early detection and referral to hospital eye services.
PURPOSE. 1.To review accuracy of referrals from the regional diabetic retinopathy screening services(DRSS) 2. To assess the appropriateness of referrals and predict the impact on hospital eye service in future.
METHODS. Retrospective data by review of patients' case notes were collected on 50 new diabetic retinopathy referrals from DRSS to the eye unit at Addenbrooke's Hospital. The visual acuities, retinopathy grading, and clinical findings recorded by DRSS were compared to ophthalmologists.
RESULTS. Of the 50, 4 did not attend their appointment. Data were collected for 46 patients over a 3 month period (July to Sep 2009).The mean age was 62 years. The visual acuities measured comparable in 100%. With regards to retinopathy grading, 50% graded as R1,30% R2,15% R3,60% M0 and 15% M1, concurred with ophthalmologists' assessment. 15 patients had “ungradable” retinopathy due to media opacity.16 patients were referred for cataract assessment. Laser photocoagulation was required by 16% of which 75% was for clinically significant macular oedema. 50% for cataract were listed for surgery. 92% were given further follow-up appointments and 8% were referred back to the community.
CONCLUSIONS. This analysis shows there is an increased demand for hospital-based assessment .The accuracy of grading ranged from 15-60%. Only a minority of patients required treatment with laser. With the incidence of diabetes rising, an efficient screening service is required to optimise community care, with hospital services focused on laser/surgical management and follow-up of high risk group. An ongoing review required for future planning and cost-structuring.

E. Dervan1, T. Wall1, J. Smith2,3
Department of Ophthalmology, Mater Misericordiae University Hospital, Dublin – Ireland
Ireland Health Service Executive Northwest – Ireland
Foresight Eye Care, Dundalk - Ireland

INTRODUCTION. We previously designed and validated a table based on the patient's pupil size and age that enables mydriasis to be targeted at those that require it in order to achieve gradable diabetic retinopathy screening photographs using a digital 45-degree non-mydriatic camera.
PURPOSE. The aim of our study was to compare the effect on the screening time and the gradability of the images produced between the standard mydriatic model where all patients are dilated and our targeted mydriasis based model.
METHODS. The participants involved consecutive patients attending retinopathy screening clinics at GP practices in Ireland in May 2009. Patients in the first 5 clinics were assigned to the standard mydriatic model and in the remaining 5 clinics to the targeted mydriasis model. The time taken to complete the screening process, the gradability of the images produced and by extrapolation the predictive accuracy of the table was measured.
RESULTS. The table successfully predicted the requirement for mydriasis in 92% of patients. The average time taken to screen patients in routine mydriasis vs. targeted mydriasis was 22.64 ± 8.13 min vs. 12.84 ± 10.81 min (P ≤ 0.0001). 97% of photographs were gradable in the routine mydriasis group compared to 94% in the targeted mydriasis group.
CONCLUSIONS. Implementation of targeted mydriasis offers the possibility of significant time efficiencies for both patient and clinic in a systematic screening programme without a significant reduction in the proportion of gradable images being acquired.

I. Ssi-Yan-Kai, A. Smith
Prince Charles Eye Unit, King Edward VII Hospital, Windsor, SL4 3DP – United Kingdom

INTRODUCTION. The UK diabetic retinopathy screening programme (DESS) is a great tool. The aim is to reduce the number of cases of blindness due to diabetic retinopathy. The diabetic screeners and graders are highly skilled at detecting and assessing diabetic retinopathy findings.
However the hospital service is becoming overwhelmed by the increase in patient referrals. This study addresses potential refinements that could help.
PURPOSE. To identify R2 referrals and determine the spectrum of diabetic eye disease.
METHODS. Data on all diabetic patients referred at King Edwards VII hospital, Windsor, by DESS for “R2, MO” staging of diabetic retinopathy were reviewed. All patients' notes, reported in the medical software MEDISOFT, between February 2009 and February 2010, were analysed. An automatic re-classification based on the ophthalmologist examination was performed. Classifications were compared and sub-groups identified.
RESULTS. Among 941 diabetic patients referred, 140 had at least one eye staged R2, MO (both eyes < R3) by DESS. All R2 DESS staging at referral were R2 after examination. Among them, 79% were staged moderate NPDR by the ETDRS and 21% were staged severe NPDR.
CONCLUSIONS. The R2 classification encompasses a range of diabetic retinopathy with different levels of severity. Some will still warrant annual review while others may imminently progress and require laser treatment. The subdivision of the DESS R2 group into 2 sub-groups may be of value in the management of these patients.

A. Sellman1, M. Löndahl2, S. Andreasson3, P. Katzman2
1Dept of Ophthalmology, Helsingborg Hospital, Helsingborg - Sweden
2Dept of Endocrinology, Skane University Hospital, Lund - Sweden
3Dept of Ophthalmology, Skane University Hospital, Lund - Sweden

INTRODUCTION. Micro, as well as macrovascular complications are more prevalent in diabetic patients with chronic foot ulcers. Retinopathy is considered to be a microvascular complication and diabetic macular oedema is an important risk factor for visual impairment.
PURPOSE. To evaluate if arterial toe blood pressure (TBP) or transcutaneous oximetry (PtcO2) in the foot could predict increased macular thickness.
METHODS. Macular thickness was prospectively measured using optical coherence tomography technique. Patients were classified as ischemic or non-ischemic according to PtcO2 at dorsum of the foot (predefined cutoff level ≤55 mm Hg) and TBI (predefined cutoff level ≤70 mm Hg).
RESULTS. Twenty patients (40 eyes) with a median diabetes duration of 22 years and an A1c of 6,35 % were included. All patients had at least one foot ulcer with a duration of at least three months despite optimal treatment at a diabetes foot clinic. Median PtcO2 was 45 mm Hg and median TBP 55 mm Hg. 14 and 13 patients were considered ischemic according to PtcO2 and TBP measurements respectively. Macular thickness was 225 (204; 250) (Q1;Q3) microns in the ischemic PtcO2 group as compared to 204 (179;223) microns in the non-ischemic (p=0,036)and 218(203;241) microns in the ischemic TBP group as compared to 197 (171;269) microns in the non-ischemic group (p=0,41).
CONCLUSIONS. PtcO2, but not TBP might be a predictor of increased macular thickness in patients with chronic diabetic foot ulcers.

B. Dupas1, S. Feldman-Billard2, E. Bui Quoc3, A. Erginay1, T. Meas4, P.J. Guillausseau4, P. Massin1
1Service d'Ophtalmologie, Hôpital Lariboisière, Paris - France
2Service de Médecine Interne, CHNO des Quinze-Vingts, Paris - France
3Service d'Ophtalmologie, Hôpital Robert Debré, Paris - France
4Service de Médecine Interne, Hôpital Lariboisière, Paris - France

INTRODUCTION. Mid-term variations of diabetic macular oedema (DME) and their causes have been poorly investigated.
PURPOSE. To describe spontaneous variations of DME over a three-month period, and to correlate the macular thickness changes with variations of blood pressure and blood glucose.
METHODS. 23 diabetic patients presenting a central macular thickness (CMT) greater than 260 µm on Optical Coherence Tomography were followed for three months. Blood pressure (BP) and CMT were measured simultaneously every two weeks. 24-hour BP monitoring and glucose monitoring were also performed during one visit.
RESULTS. The mean variation of CMT was 88±71.4 μm [range 21-356]. Twelve of 23 patients presented at least once a variation of their CMT greater than 12% between two visits. The coefficient of cumulative variation (CCV) of parameters was defined as the sum of the variations of absolute values of the parameter between each visit divided by the number of visits. The 12 patients with significant CMT variations had a CCV of systolic and diastolic BP twice as high as the 11 patients with a non significant variation of CMT. They also presented the higher nocturnal BP variability and the higher intraday glucose excursions.
CONCLUSIONS. The high variability of macular thickness should be taken into consideration by practitioners, and justifies a control group in every study investigating the benefit of a therapy in DME. Further studies are needed in order to understand the role of systemic factors on DME.

E. Provins, S. Craske, S. Chave, I.M. Stratton, C. Martin, P.H. Scanlon
English National Screening Programme for Diabetic Retinopathy, Cheltenham – United Kingdom

INTRODUCTION. Gloucestershire Diabetic Retinopathy Screening Service (DRSS) and General Practices (GP) maintain screening lists with a manual process to identify patients with diabetes to be called for annual screening.
PURPOSE. Increasing diabetes incidence led Gloucestershire DRSS to compare a list from electronic GP patient records with the manual list.
METHODS. On 17 April 2009 an electronic list of people with diabetes at each GP was created.
New patients were identified by comparing NHS Numbers on electronic list with Gloucestershire DRSS Orion Patient Management System. New patients identified were investigated.
RESULTS. This pilot identified 14,919 patients aged 12 years or above from 54 of 85 Gloucestershire GP Practices. Of these 712 patients (4.8% increase) were not previously known to the DRSS. Of these, 342 were diagnosed with diabetes within 90 days of data extraction, 23 had recently moved into the area and 17 had reached 12 years of age more than 90 days before. Of those not previously known to DRSS 121 were known to have been diagnosed with diabetes for more than 1 year. 69 were not eligible for screening (25 incorrectly coded, 34 GP exclusions, 4 deceased). Of 267 newly identified and screened for the first time 96 had DR, of which 13 were referred to the hospital eye service.
CONCLUSIONS. A pilot project of electronic transfer of data to Gloucestershire DRSS from GPs identified substantial numbers of diabetic patients. Many young patients had not been transferred to DRSS at age 12 years.

I. Klaassen1, J.M. Hughes1, W. Kamphuis2, C.J.F. Van Noorde1, R.O. Schlingemann1
1Ocular Angiogenesis Group, Departments of Ophthalmology and Cell Biology and Histology, Academic Medical Center, University of Amsterdam, Amsterdam - The Netherlands
2Astrocyte Biology & Neurodegeneration, Netherlands Institute for Neuroscience, Amsterdam - The Netherlands

INTRODUCTION. Real-time quantitative PCR (qPCR) is becoming a popular method for quantification of specific mRNA transcript levels in tissues. Because of its high sensitivity, the interpretation of qPCR data relies heavily on normalization strategies in order to correct for sampling errors, qPCR efficiency variations and other sources of variation.
PURPOSE. We investigated a new normalization method without the use of reference genes, since many of these have been reported to be regulated under diabetic conditions.
METHODS. On 33 samples of STZ-induced diabetic rat retina we employed a normalization procedure that is based on total amount of synthesised first strand complementary DNA (cDNA) and compared this approach with conventional normalization strategies using reference genes. The expression profiles of seven commonly used reference genes and four retina specific endogenous control genes were analyzed and gene stability was assessed with normalization algorithms geNorm and NormFinder.
RESULTS. When different normalization methods were tested on the target gene Icam-1, gene expression levels were significantly affected by the reference gene or normalization method used. Validation of all normalization methods suggested that the total amount of synthesised first strand cDNA is the most appropriate method for use in our samples.
CONCLUSIONS. This study shows the difficulties and pitfalls of finding suitable reference genes in diabetic rat retina. We developed a method that is rapid, economical and can be used under all conditions including complex diseases such as diabetes.

I.M.C. Vogels, I. Klaassen
Ocular Angiogenesis Group, Departments of Ophthalmology and Cell Biology and Histology, Academic Medical Center, University of Amsterdam, Amsterdam - The Netherlands

INTRODUCTION. In the eye, structural microvascular changes in retina are the main cause of diabetic retinopathies and blindness.
PURPOSE. Our aim was to develop a method to isolate and culture retinal endothelial cells from diabetic and non-diabetic donors in order to study their differences in growth and response to experimental diabetic conditions.
METHODS. Freshly enucleated donor eyes (up to 24 hours post mortem) were obtained from the Corneabank Amsterdam (The Netherlands) after removal of corneal buttons for transplantation. Retinas were homogenised in DMEM by mechanical dissociation. Endothelial cells were isolated by enzymatic digestion and differential filtration and grown in complete ECM2 culture medium on fibronectin and collagen coated culture plates. Using this method, cells can be isolated within 2 hours.
RESULTS. The human retinal endothelial cell cultures exhibited the characteristic “cobblestone” growth pattern as seen by phase contrast microscopy. By using immunohistochemistry techniques and RT-PCR the cells were identified as endothelial cells by the presence of von Willebrand factor, CD31, vascular endothelial cadherin and claudin-5. The cultures were not contaminated by pericytes as demonstrated by the lack of immunoreactivity of NG2.
CONCLUSIONS. This method is fast and reproducible and makes it possible to study differences between diabetic and non-diabetic human retinal endothelial cells in primary cultures.

M. Karthikeyan1, H. Cilliers2, L. P. Chow3, D. Verma1
1Addenbrooke's University Hospital, Cambridge University Hospitals NHS Foundation Trust, Cambridge – United Kingdom
2Cheltenham General Hospital, Gloucestershire Hospitals NHS Foundation Trust – United Kingdom
3John Radcliffe Hospital, Oxford Radcliffe Hospitals NHS Trust, Oxford – United Kingdom

INTRODUCTION. The SEA-DRSS refer mainly to AUH-DRS. To better monitor outcomes, a database of referrals was created aimed to establish a network between SEA-DRSS and AUH-DRS to share and correlate data.
PURPOSE. 1. To assess referrals from SEA-DRSS to AUH-DRS and once assessed to monitor outcomes. 2. Are NICE referral guidelines being met and to predict impact on AUH-DRS.
METHODS. Prospective data of referrals and review of 50 patients in database from July–September 2009. All referral data recorded at hospital assessment: visual acuities, ocular and systemic co-morbidity, BP, HBA1c, grading according to DRS &EDTRS, outcomes for treatment and follow-up.
RESULTS. Data were collected for 46 patients with 3 (6%) non-attendance. Male:female=1.2: 1 DM Type1:Type2=1:8 Mean age 26yrs DMType1; 64yrs for DMType2. 9.6w between referral and assessment. Meeting NICE guidelines and DR progression due to delay will be discussed. Correlation of VA and grading data between SEA-DRSS &AUH-DRS will be compared by statistical analysis taking contributing factors for DR like BP &HBA1c into account. Laser required by 16% with treatment and frequency will be discussed. Of 16 (32%) referred for cataracts 50% were listed for surgery. 92% required follow-up and we plan to elaborate on the impact on AUH-DRS.
CONCLUSIONS. Increased demand is generated by referrals. Though few required laser, there is significant follow-up required. Efficient and accurate DRSS-referrals need to be matched in assessment by excellent hospital-DRS with capacity to provide laser/surgical treatments and follow-up. Prospective review of referrals and outcomes is required for future planning and cost-structuring of AUH-DRS.

I. Leung, H. Kuper, T. Peto, W. Mathenge
Moorfields Eye Hospital NHS Foundation Trust London School of Hygiene and Tropical Medicine

INTRODUCTION. There is little information available on the severity of diabetes related eye diseases in Kenya.
PURPOSE. The purpose of this study was to evaluate the use of digital colour fundus photographs in Kenya for the severity of diabetic retinopathy based on the UK National Guidelines on Screening and Grading for Diabetic Retinopathy.
METHODS. Colour fundus images were obtained using Topcon NW6S non-mydriatic digital retinal camera in a randomly selected representative population of Kenya. A 45 degree field was used to capture two images per eye; one is fovea centred and one is disc centred in 3460 patients (6920 eyes). All images were graded in Adobe Photoshop grading for the abnormalities listed in the UK National Guidelines on Screening for Diabetic Retinopathy (DR).
RESULTS. Altogether, 3460 patients had photographs taken. Of these, 6437 eyes were gradable, 530 displayed signs of diabetic retinopathy or maculopathy. 360 eyes had mild, 92 had moderate, 30 had severe and 48 had proliferative DR. For maculopathy, 392 eyes had non-sight threatening, and 12 had sight threatening maculopathy, in 3 the level of maculopathy was impossible to decide due to photo quality. Based on clinical data, 6.5% of the patients had diabetes and a third of these showed signs of diabetic retinopathy/maculopathy.
CONCLUSIONS. DR and maculopathy are not uncommon in Kenya. The results of this survey could be used to guide further planning of eye care provision for patients with DR.

T. Peto, H. Abdillahi, F. Sallo, I. Leung and MacTel Study Group
Moorfields Eye Hospital NHS Foundation Trust - London
MacTel Study Group

INTRODUCTION. Type 2 macular telangiectasia (MacTel) is a bilateral disease affecting central vision. The MacTel Project enrolls patients from 26 sites around the world; those in the study have multiple imaging performed yearly.
PURPOSE. The aim was to establish the severity and the progression of diabetic retinopathy (DR) in MacTel.
METHODS. Colour fundus, fluorescein angiographic, OCT and autofluorescence images of MacTel Study patients are graded at the Reading Centre of Moorfields Eye Hospital, UK for characteristics of MacTel; then diabetes status is obtained from the co-ordinating centre. Of those with diabetes, 7-field stereo images for the baseline, and then 3-fields for follow-ups are graded using ETDRS standards.
RESULTS. The mean age of MacTel patients at diagnosis was 57±9 years. Out of 409 patients enrolled, 124 had known diabetes. Of these, 96 patients had no diabetic retinopathy or maculopathy at baseline, and only 5 had Level 35 disease in at least one eye. 68/96 remained without DR at 12 months, and 26 are still without DR at 36 months follow-up. Of those who progressed, one developed mild and one moderate non-proliferative DR. No patient developed proliferative DR during the follow-up period. Only one patient had diabetic maculopathy requiring no laser treatment.
CONCLUSIONS. Diabetes mellitus in MacTel is a common finding, but DR does not seem to progress to sight-threatening disease. It might warrant further exploration whether these patients show differences in the clinical characteristics of diabetes compared to patients with diabetes who exhibit no signs of MacTel.

  « previous page    



W3C standards logos

site developed by Netsima Ltd »