Association of retinal vessel calibre with diabetic retinopathy in an urban Australian Indigenous population *

Transcription

1 Original Article Clinical Science Association of retinal vessel calibre with diabetic retinopathy in an urban Australian Indigenous population * Mohamed Dirani PhD, 1 Annie K McAuley BSc, 1 Louise Maple-Brown PhD, 3,4 Ryo Kawasaki PhD, 1,2 RL McIntosh MSc, 1 C Alex Harper MD, 1 Ecosse L Lamoureux PhD, 1 Shaun Tatipata PhD, 5 Terry Dunbar PhD, 6 Kerin O Dea PhD 3,7 and Joan Cunningham PhD 3 1. Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, University of Melbourne, Australia 2. Centre for Clinical Research Excellence in Clinical Research in Diabetes, St. Vincent s Hospital, Australia 3. Menzies School of Health Research, Institute of Advanced Studies, Charles Darwin University, NT, Australia 4. Division of Medicine, Royal Darwin Hospital, NT, Australia 5. Danila Dilba Health Services, Darwin, NT, Australia 6. School of Education, Faculty of Education, Health and Science, Charles Darwin University, NT, Australia 7. Sansom Institute for Health Research, University of South Australia, SA, Australia Corresponding author: Mohamed Dirani, PhD Centre for Eye Research Australia 32 Gisborne Street, East Melbourne 3002 Victoria, Australia dirani@unimelb.edu.au This is an Accepted Article that has been peer-reviewed and approved for publication in the Clinical and Experimental Ophthalmology, but has yet to undergo copy-editing and proof correction. Please cite this article as an "Accepted Article"; doi: /j x

2 Short running title: Retinal Vessel Calibre with Diabetic Retinopathy Received 20 January 2010; accepted 1 March 2010

3 ABSTRACT Objective: To assess the relationship of retinal vessel diameter and diabetic retinopathy in a subgroup of participants recruited through the Darwin Region Urban Indigenous Diabetes (DRUID) study. Methods: Participants were examined as part of the DRUID study. All participants with gradable fundus photographs were included in the current analysis. Assessment of retinal vascular diameter, including arteriolar diameter (central retinal arteriolar equivalent, CRAE), and venular diameter (central retinal venular equivalent, CRVE) was undertaken using a semiautomated retinal vascular imaging program. Diabetic Retinopathy (DR) was graded according to the modified Early Treatment DR Study (EDTRS) scale. Results: A total of 110 participants, 25 males and 85 females, with a mean age of 50.8 years were included in the analysis. The odds ratio for having DR for each standard deviation increase in CRVE was as high as 1.62 (95%CI 0.94, 2.80), however this did not reach statistical significance (p = 0.08). Moreover, individuals with severe Non-Proliferative DR (NPDR) and Proliferative DR (PDR) were found to have narrower arteriolar diameters compared to those with no DR, but this was not statistically significant (-8.1μm, 95%CI, -39.3μm, 23.1μm; p = 0.612). Conclusion: Our data indicates a trend for narrower arteriole diameter and wider venular diameter with DR in this high risk ethnic group, which concurs with overall trends seen in non- Indigenous populations. Keywords: diabetic retinopathy, arteriolar, venular, diabetes

4 INTRODUCTION Diabetic Retinopathy (DR) is one of the most common vascular complications of Diabetes Mellitus (DM) [1]. More than half of individuals who have suffered from DM for 20 years or more are reported to have some level of DR [2]. The prevalence of DM has been shown to vary with age, gender, ethnicity and other demographic factors, such as socio-economic status and educational attainment. For instance, in Australia alone, the prevalence of DM in Indigenous Australians has been reported to be 10 times higher compared to that in the general Australian population, with diabetic-related complications being one of the leading causes of blindness and mortality in the Indigenous Australian community [3, 4]. DR can be categorized into early stages of non-proliferative DR and a later stage of proliferative DR, with the latter typically associated with significant vision loss. Non-invasive retinal photography and recent advances in imaging technology have allowed measurements of the retinal microvasculature [5-10]. Epidemiological studies have reported significant associations between change in retinal vessel calibre with systemic conditions and their complications, including DR [11-14]. However, the exact relationship of retinal vessel diameter with DR is unclear. In the Australian Diabetes, Obesity and Lifestyle (AusDiab) study, wider retinal arteriolar calibre was associated with up to 2-fold higher likelihood of having DR in individuals with diabetes [14]. In another study that investigated incident cases of diabetes in middle aged persons (49 to 73 years), retinal arteriolar narrowing was independently associated with diabetes [15]. Nonetheless, the majority of these studies have been conducted in nonindigenous populations. Investigating the relationship between retinal vascular parameters and DR in the Australian Indigenous community is of significance considering the elevated prevalence of known diabetic risk factors (such as obesity, smoking and hypertension), and the lower age of diabetes onset [16], and the potential for the retinal vascular assessment to provide a novel method to identify persons at risk of DR. To our knowledge there are no published data on the relationship between retinal vessel calibre and DR in the Australian Indigenous population. In this paper we report on the association of specific retinal vessel calibre with DR using a sub-group from the Darwin Region Urban Indigenous Diabetes (DRUID) Study [17].

5 MATERIALS AND METHODS Participants Of 172 participants with diabetes in the DRUID Study, 136 consented to and were available for assessment of complications of diabetes. Of these, 115 had digital retinal photographs taken, and 110 subjects had suitable quality retina images for measuring retinal vessel calibres using standardized software and were included in the current analysis. The methodology of the DRUID study has been published elsewhere [17]. In brief, diabetes diagnosis was conducted using a 75 gm oral glucose tolerance test (OGTT), adhering to the World Health Organization (WHO, 1999) diabetes classification [18]. Individuals were considered as having diabetes if: (a) fasting plasma glucose was 7.0mmol/L; (b) 2 hour plasma glucose was 11.1mmol/L; or (c) they reported current use of diabetic medication(s). Ethics The DRUID study protocol [17] was approved by the joint Human Research Ethics Committee of the Menzies School of Health Research and the Northern Territory Department of Health and Community Services. Assessment of diabetic complications Retinal photography was performed by a specialist eye Aboriginal Health Worker (ST) using a Topcon non-mydriatic retinal camera (TRC-NW5S, Topcon Corporation, Japan). Approximately 50% of our participants underwent dilation, with this being achieved using 1 drop of Tropicamide (0.5%), followed by a single drop of phenylephrine (2.5%). Two standard fundus photographs were taken for each eye, including both macular and optic disc fields. Retinal images were assessed by an ophthalmologist (CAH) at the Centre for Eye Research Australia (CERA), University of Melbourne. Images were graded according to a simplified version of the Early Treatment Diabetic Retinopathy Study (EDTRS) [19] severity scales and categorized into the following clinical stages: no DR; mild or moderate non-proliferative DR (NPDR); severe NPDR and/or proliferative DR (PDR); fundus pathology other than DR-related. Retinal vessel calibre measurements

6 Retinal vascular diameters were measured using a validated computer-based program [20]. One disc-centred digitized image from each eye was used for this measurement. For each photograph, all arterioles and venules coursing through an area one-half to one disc-diameter from the optic disc were measured and calibres of the largest 6 arterioles and largest 6 venules were summarized as the central retinal artery equivalent (CRAE) and the central retinal vein equivalent (CRVE), respectively [21]. Using both CRAE and CRVE, the arteriolar venular ratio (AVR) was calculated. Statistical Analysis Analysis was conducted using generalized estimating equation models accounting for right and left eye data together. Firstly, we estimated mean difference in CRAE, CRVE and AVR by the severity of DR, adjusting for age and gender (Model 1) and adjusting for age, gender, smoking, mean arterial blood pressure, C-reactive protein (CRP), body mass index (BMI), history of laser treatment and haemoglobin A1c (HbA1c) (Model 2). Then we estimated odds ratios (OR) and their 95% confidence interval values (CI) of having DR for persons with the smallest tertile compared to persons with highest tertile in CRAE and AVR, and for persons with the highest tertile compared to persons with the smallest tertile in CRVE in multivariate models (Model 1 and Model 2). All statistical analyses were performed using Stata (version 10.1, StataCorp. College Station, TX, USA). RESULTS Demographics A total of 110 Indigenous Australians participants (25 males and 85 females), with a mean age of 50.8 years (SD: 11.1 years) were included in the current analysis. The average age at which diabetes was diagnosis was 43.7 years (SD: 12.0 years), with the mean haemoglobin A1c being 7.9% (SD: 1.8%) (Table 1). Mean systolic and diastolic BP was 127mmhg and 78mmhg, respectively (Table 1). More than one-third (37.6%) were current smokers (Table 1). Thirty participants had DR, with 26 having mild/moderate NPDR and 4 having severe NPDR or PDR.

7 Retinal Parameters Using Model 1, individuals with any DR had a significantly lower mean AVR (-0.03, 95%CI -0.07, ; p = 0.041) compared to that of the reference group, after adjusting for age and gender (Table 2). However, after adjusting for all covariates (Model 2), the statistical significance diminished (p = 0.073). Moreover, no significant differences for CRAE and CRVE measurements were found using Model 1 (Table 2). For Model 2, the mean difference for CRAE measurements between those with severe NPDR/PDR and those with no DR (reference group) was as high as 8.1µm, but this did not reach statistical significance (p = 0.612). No significant mean differences for CRVE measurements were found between any DR group and the reference group for both models. Using both multivariate models, no significant association was found between DR and CRAE and CRVE measurements after adjusting for age, gender, smoking, mean arterial BP, CRP, BMI, history of laser treatment and HbA1c (Table 3). Moreover, no significant association was evident for the presence of DR with each SD change in CRAE and CRVE, however it approached statistical significance for each SD change in CRVE (OR: 1.62, 95% CI: , p = 0.08). DISCUSSION We explored the relationship of retinal vascular parameters and DR in a subgroup of participants recruited through the Darwin Region Urban Indigenous Diabetes (DRUID) Study [17]. This is the first study to describe the relationship between the retinal microvasculature and diabetic retinopathy in Indigenous Australians. Although we have a relatively small sample size, our findings of a trend towards larger CRVE with DR are consistent with that of previous findings in different populations [10, 14, 22, 23]. However, contrary to previous findings [12, 24, 25], in our sample, Indigenous Australians with DR had narrower arteries compared to those without DR. There is a lack of consensus in the literature about the association of wider retinal arteriolar calibre and the risks of DR. Two population-based prevalence studies, the Multi-Ethnic

8 Study of Atherosclerosis (MESA) and the Wisconsin Epidemiological Study of Diabetic Retinopathy (WESDR), reported a lack of association between CRAE and DR [10, 22, 23]. On the other hand, other studies have shown that DR is associated with enlarged retinal arteriolar calibre [12, 24, 25]. The AusDiab study, for example, reported that wider retinal arteriolar calibre was independently associated with increased risks of DR, after adjusting for BP, glycemic control and other retinopathy risk factors [12]. Similarly, a study of Australian children and adolescents with type 1 diabetes also found larger arteriolar calibre is predictive of incident DR, independent of other retinopathy risk factors. Our study found that narrower arterioles were significantly associated with DR, but the statistical significance did not hold after adjustment for other relevant risk factors. Several mechanisms could explain the discrepancies between our study findings and others. Firstly, compared to previous studies [10, 12, 22, 23], our sample size was relatively small (n = 110 individuals). Secondly, there was a significantly greater use of cardiovascular medications in the group with DR (statins and ACE inhibitors or angiotensin-2 receptor antagonists). Thirdly, it is also possible that there is a unique diabetes disease process in the Indigenous Australians compared to other populations. However, a substantially larger sample size would be required to assess this possibility. Fourthly, we cannot rule out the possibility that attenuation of retinal vessels after pan-retinal laser photocoagulation in advanced DR might be over-represented in our small sample. Finally, the statistical power of our study is quite restricted due to the adjustment of a large number of confounding factors with the use of a small sample. Therefore, our study findings should be considered in light of these limitations. The strengths of our study include the clinical measurements of DR and retinal parameters, the use of sophisticated software to quantify retinal calibre conducted at a specialist grading centre and the participation of a unique and high risk Australian population group. However, our small sample size is a serious limitation and would have significantly reduced the statistical power of the study analysis. In addition, it is worth noting that only 50% of our participants had dilated fundus photography. Nonetheless, the current study has provided insights into the relationship of microvascular changes with DR in the Indigenous Australian community. This study could pave the way for future studies into understanding the differences in disease patho-physiology (including macular edema) between different populations

9 particularly that in the Indigenous community where premature mortality related to diabetes remains to be of profound concern [26]. Further study is also needed to assess the utility of this technique as a useful screening tool in this high risk population. Acknowledgements The authors gratefully acknowledge the support of DRUID study participants, study staff, members of the Indigenous Steering Group, and partner organisations. The DRUID Study was funded by the National Health and Medical Research Council (NHMRC Project Grant #236207), with additional support from the Australian Government Department of Employment and Workplace Relations, the Clive and Vera Ramaciotti Foundation, the Vincent Fairfax Family Foundation, the International Diabetes Institute (AusDiab Partnership), and Bayer HealthCare. The DRUID Study is an in- kind project of the Cooperative Research Centre for Aboriginal Health. LMB had a NHMRC Scholarship and is supported by the Centre of Clinical Research Excellence in Clinical Science in Diabetes, University of Melbourne. We acknowledge the NHMRC funding for Diabetes Centre for Clinical Research Excellence (CCRE) and the kind contributions from the Menzies Foundation. JC was supported by a NHMRC Career Development Award (#283310) and NHMRC Research Fellowship (#545200). Funding sources played no role in the study design, in the collection, analysis and interpretation of the data, in the writing of the manuscript, or in the decision to submit the manuscript for publication.

10 REFERENCES 1. Mohamed, Q., M.C. Gillies, and T.Y. Wong, Management of diabetic retinopathy: a systematic review. JAMA, (8): p Tapp, R.J., et al., The prevalence of and factors associated with diabetic retinopathy in the Australian population. Diabetes Care, (6): p The Health and Welfare of Australia's Aboriginal and Torres Strait Islander Peoples. 2005, Australian Bureau of Statistics and Australian Institute of Health and Welfare, ABS & AIHW: Canberra. 4. K. O'Dea, M. Patel, and D.C.-. D. Kubisch, Obesity, diabetes, and hyperlipidemia in a central Australian aboriginal community with a long history of acculturation. Diabetes Care, : p Couper, D.J., et al., Reliability of retinal photography in the assessment of retinal microvascular characteristics: the Atherosclerosis Risk in Communities Study. Am J Ophthalmol, (1): p Sherry, L.M., et al., Reliability of computer-assisted retinal vessel measurementin a population. Clin Experiment Ophthalmol, (3): p Hubbard, L.D., et al., Methods for evaluation of retinal microvascular abnormalities associated with hypertension/sclerosis in the Atherosclerosis Risk in Communities Study. Ophthalmology, (12): p Li, H., et al., Automatic grading of retinal vessel caliber. IEEE Trans Biomed Eng, (7): p Wong, T.Y., et al., Computer-assisted measurement of retinal vessel diameters in the Beaver Dam Eye Study: methodology, correlation between eyes, and effect of refractive errors. Ophthalmology, (6): p Wong, T.Y., et al., Retinal vascular caliber, cardiovascular risk factors, and inflammation: the multi-ethnic study of atherosclerosis (MESA). Invest Ophthalmol Vis Sci, (6): p Nguyen TT, et al., Relationship of retinal vascular caliber with diabetes and retinopathy: the Multi-Ethnic Study of Atherosclerosis (MESA). Diabetes Care, : p

11 12. Rogers SL, et al., Retinal arteriolar caliber predicts incident retinopathy: the Australian Diabetes, Obesity and Lifestyle (AusDiab) study. Diabetes Care., : p Klein R, et al., Retinal vascular caliber in persons with type 2 diabetes: Epidemiological Study of Diabetic Retinopathy. Ophthalmology, : p Tikellis G, et al., The relationship of retinal vascular calibre to diabetes and retinopathy: the Australian Diabetes, Obesity and Lifestyle (AusDiab) study. Diabetologia, : p Wong TY, et al., Retinal arteriolar narrowing and risk of diabetes mellitus in middleaged persons. JAMA, : p Bruce DG, et al., Diabetes education and knowledge in patients with type 2 diabetes from the community: the Fremantle Diabetes Study. J Diabetes Complications, : p Cunningham J, et al., Study protocol--diabetes and related conditions in urban indigenous people in the Darwin, Australia region: aims, methods and participation in the DRUID Study. BMC Public Health, : p. 6: World Health Organization: Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation. Part 1: Diagnosis and Classification of Diabetes Mellitus , World Health Org.: Geneva. 19. Cantrill HL, The diabetic retinopathy study and the early treatment diabetic retinopathy study. Int Ophthalmol Clin., : p Hubbard LD, et al., Methods for evaluation of retinal microvascular abnormalities associated with hypertension/sclerosis in the Atherosclerosis Risk in Communities Study. Ophthalmology, : p Knudtson MD, et al., Revised formulas for summarizing retinal vessel diameters. Curr Eye Res, : p Klein, R., et al., Are inflammatory factors related to retinal vessel caliber? The Beaver Dam Eye Study. Arch Ophthalmol, (1): p Klein, R., et al., Are retinal arteriolar abnormalities related to atherosclerosis?: The Atherosclerosis Risk in Communities Study. Arterioscler Thromb Vasc Biol, (6): p

12 24. Stefansson E, Ocular oxygenation and the treatment of diabetic retinopathy. Br J Ophthalmol : p Frederiksen C, et al., The blood pressure-induced diameter response of retinal arterioles decreases with increasing diabetic maculopathy. Graefes Arch Clin Exp Ophthalmol, : p O'Dea K, et al., Diabetes and cardiovascular risk factors in urban Indigenous adults: Results from the DRUID study. Diabetes Res Clin Pract., : p

13 Table 1: The socio demographic, personal and clinical characteristics of our study participants Total (n=110) Mean (Standard Deviation) or n(%) Age (years) 50.8 (11.1) Female n (%) 85 (77.3) Systolic BP (mmhg) (17.5) Diastolic BP (mmhg) 78.1(9.8) Hypertension n (%) 56 (50.9) Total Cholesterol (mmol/l) 5.31 (1.40) Triglycerides (mmol/l) 2.1 ( )* HDL Cholesterol (mmol/l) 1.00 (0.31) LDL Cholesterol (mmol/l) 3.04 (1.06) Body Mass Index (kg/m 2 ) 31.9 (6.5) Haemoglobin A1c (%) 7.9 (1.8) Insulin treated n (%) 16 (14.5) High sensitivity-crp (mg/l) 7.2 ( )* Age at diagnosis (years) 43.7 (12.0) Current Smoker n (%) 41 (37.3) BP: Blood pressure; HDL: high density lipoprotein; LDL: Low density lipoprotein; CRP: C- reactive protein * Median and inter-quartile range.

14 Table 2: Vessel Calibre Measurements by Diabetic Retinopathy Status (Based on both eyes data in GEE models) Model 1 Model 2 Mean (SD) Mean Diff. Mean Diff. 95%CI p-value (μm) (μm) (μm) 95%CI p-value CRAE No DR (n=179 eyes) (14.8) (Reference) - - (Reference) - - Any DR (n=30 eyes) (17.6) -3.7 (-10.0, 2.6) (-9.63, 6.89) Mild/Mod NPDR (n=24 eyes) (17.1) -2.4 (-9.2, 4.4) (-8.74, 7.15) Severe NPDR/PDR (n=6 eyes) (21.7) (-25.6, 4.7) (-39.3, 23.1) CRVE No DR (24.7) (Reference) - - (Reference) - - Any DR (30.0) 6.9 (-3.5, 17.3) (-2.11, 22.8) Mild/Mod NPDR (29.9) 6.5 (-4.8, 17.7) (-1.87, 23.2) Severe NPDR/PDR (34.8) 9.4 (-15.6, 34.4) (-46.8, 48.9) AVR No DR 0.69 (0.08) (Reference) - - (Reference) - - Any DR 0.66 (0.10) (-0.07, ) (-0.08, 0.004) Mild/Mod NPDR 0.67 (0.10) (-0.06, 0.01) (-0.08, 0.01) Severe NPDR/PDR 0.62 (0.07) (-0.15, 0.01) (-0.24, 0.09) Model 1: Adjusted for age and gender; Model 2: Adjusted for age, gender, smoking, mean arterial blood pressure, CRP, BMI, history of laser treatment and HbA1c.

15 Table 3: Risk of having DR by tertiles or 1 SD change of retinal vessel calibres Prevalence of DR Model 1 Model 2 n (%) OR 95%CI p-value OR 95%CI p-value CRAE Tertile 1 10 (14.9%) 1.50 (0.53, 4.23) (0.17, 3.05) 0.65 Tertile 2 9 (13.0%) 1.34 (0.47, 3.85) (0.28, 4.19) 0.90 Tertile 3 7 (10.1%) Per - 1SD (0.84, 1.92) (0.54, 1.54) 0.74 CRVE Tertile 1 9 (13.0%) Tertile 2 5 (7.4%) 0.56 (0.17, 1.80) (0.06,2.58) 0.34 Tertile 3 12 (17.7%) 1.52 (0.59, 3.92) (0.53, 7.34) 0.31 Per +1SD (0.86, 2.04) (0.94, 2.80) 0.08 AVR Tertile 1 9 (16.7%) 2.35 (0.77, 7.13) (0.44, 7.24) 0.42 Tertile 2 11 (13.8%) 1.88 (0.65, 5.44) (0.41, 7.13) 0.46 Tertile 3 6 (8.5%) Per - 1SD (0.99, 2.43) (0.88, 2.43) 0.14 Model 1: Adjusted for age and gender; Model 2: Adjusted for age, gender, smoking, mean arterial blood pressure, CRP, BMI, history of laser treatment and HbA1c. CRAE: Central Retinal Artery Equivalent; CRVE: Central retinal vein equivalent; DR: Diabetic retinopathy; OR: Odds ratio; CI: Confidence interval; SD: Standard deviation N = 26 individuals