Ophthalmology (Eye & Ear Hospital) - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
ItemOptimising the management of diabetic retinopathy in pregnant women with pre-existing diabetes in Australia( 2021)Background: Diabetic Retinopathy (DR) is a leading cause of blindness among people of reproductive age and is thought to be worsened by pregnancy. Findings from prior studies have been conflicting, many are outdated, and the majority only studied women with type 1 diabetes (T1DM). It is crucial to identify those within this population who are most at risk of developing DR progression towards sight-threatening disease, so that appropriate management can be planned. Clinical decisions regarding DR management in pregnancy can be fraught, due to the need to give timely sight-saving treatment while also minimising adverse effects to the foetus. Despite these concerns, DR management guidelines have been inconsistent across the globe due to disparate findings. Purpose: This thesis aimed to 1) report the prevalence of DR, sight-threatening DR (STDR), and its progression rate during pregnancy in Metropolitan Melbourne; 2) assess the use of optical coherence tomography angiography (OCTA) in evaluating retinal changes in diabetic patients during different stages of pregnancy; and 3) determine the adherence rate and the barriers to the recommended eye screening guidelines in pregnant women with pre-existing diabetes. Research Design and Methods: This doctoral research project was a prospective longitudinal cohort study of women with T1DM or type 2 diabetes who were pregnant or planning to get pregnant from two tertiary maternity hospitals in Melbourne, Australia. Eye examinations were scheduled in each trimester and at 3-months postpartum. DR severity was graded for each eye from 2-field retinal photographs. Progression was defined as worsening by >= 1-step on the Airlie House classification, development of diabetic macular oedema (DMO), or the need for laser treatment during pregnancy. Sight-threatening (ST) progression was defined as development of proliferative DR or DMO. Additionally, 3x3-mm OCTA scans (macula-centred) were taken from participants who attended Melbourne Eyecare Clinic using swept-source OCT (Triton, Topcon Corp). Foveal avascular zone (FAZ) area and vessel density (VD) from the superficial capillary plexus were measured. Barriers to attending eye screening were assessed with the modified Compliance with Annual Diabetic Eye Exams Survey. Results: Aim 1: A total of 147 from 191 eligible women (77%) were recruited, with at least one eye exam performed in 130 (88.4%). Sixty-two women (47.7%) had T1DM. DR and STDR prevalence were 20.8 (95% confidence interval [CI] 16.3-26.1) and 6.6 (CI 4.1 - 10.4) per 100 eyes, respectively. Among the 144 eyes (72 women) with more than one eye exam, 9.7% (CI 5.8 - 15.8) had DR progression and 4.5% (CI 1.9 - 9.6) had ST progression. Elevated systolic blood pressure (risk ratio [RR] 10.36, CI 3.14 - 34.12) and pre-existing DR (RR 5.07, CI 1.90 - 13.49) in early pregnancy significantly increased the risk of progression. Aim 2: A total of 125 eyes from 64 women were imaged. The majority of women were in the pregnant group (77%) with 38 (59%) having T1DM. The majority of eyes had no DR (81.6%) or DMO (97.6%). The VD in the pericentral region and superior, temporal, and nasal subfields were significantly lower in eyes with DR than eyes without in the non-pregnant and pregnant groups (p-values <=0.039). The nasal VD in eyes with no DR in the pregnant group was significantly higher than that of the postpartum group (p=0.010). Although not reaching statistical significance, a trend towards higher VD in the pregnant group than the non-pregnant group was observed in all subfields, except in the temporal subfield. The temporal VD was lower in the pregnant group than the non-pregnant group and was especially marked in eyes with DR. No significant difference was found in the mean FAZ area between various pregnancy statuses and DR severities. Aim 3: There were 125 pregnant women who completed the survey. The median age was 34 years (range 19-47). Sixty-four respondents (51.2%) had T1DM. A retinal assessment was performed at least once in the first trimester in 56 (44.8%) and 33 (26.4%) had the ideal number of examinations. Competing priorities (p=0.012) and the belief of having reasonable diabetes control (p=0.007) were the main reasons given for non-attendance. Interestingly, the presence of multiple eye care providers in the neighbourhood was also associated with non-adherence (p=0.003). Conclusion: This PhD study has provided updated data on the prevalence of DR and its rate of progression in the Australian pregnant population with pre-existing diabetes, illustrated the use of OCTA in this population, and reported the adherence rate and barriers to the recommended eye screening guidelines. The DR prevalence in pregnant women was similar to the non-pregnant diabetic population in Australia. However, nearly 1 in 10 eyes had DR progression during pregnancy, with almost half of these developing STDR. DR presence could be reflected by a decrease in the VD in the OCTA scan, which is in concordance with the findings from previous studies in non-pregnant populations, suggesting a potential role for OCTA in the evaluation of DR during pregnancy in a non-invasive manner. Worryingly, more than half of our cohort did not adhere to the recommended eye screening schedule in pregnancy, and about 70% did not receive frequent exams as recommended, highlighting the need to address barriers to eye screening attendance given the significant risk of vision loss from DR in this population. This study benefits society by providing findings that may help inform the future development of evidence-based guidelines for DR screening and management in this unique and growing population. Based on this study’s findings, it is recommended that pregnant women with pre-existing diabetes should have an eye examination in the first trimester or as soon as pregnancy is recognised. If no DR is found, an additional exam during pregnancy is advised only for women with T1DM. However, if DR is present, an eye exam in each trimester is recommended irrespective of diabetes type.
-
ItemBiomarkers in diabetic retinopathy: genetic and proteomic profiling( 2014)Diabetic retinopathy (DR) is a common complication of diabetes mellitus and currently threatens the vision of over 100 million people worldwide. Blindness in diabetes is mostly preventable if sight threatening DR (STDR) is detected and treated early. However, the ophthalmic course in people diagnosed with diabetes can be difficult to predict, as risk indicators are not DR specific. This study adopted a multifactorial approach, assessing systemic circulating biomarkers and genetic markers, to investigate risk of blinding disease. The overriding aim of this thesis was to identify protein and genetic blood-based biomarkers that are consistently associated with severe STDR. METHOD: Samples were obtained from the Diabetes Management Project (DMP). The proteomic fraction of the thesis comprised 190 plasma samples from Type 2 diabetic patients, (n=60) STDR (proliferative DR (PDR) or severe non PDR (NPDR), (n=60) minimal or moderate NPDR, patients without DR (n=60) and (n=10) reference sample. Protein levels were analysed using iTRAQ conducted using mass spectrometry. To replicate this work a total of 70 samples with Type 2 diabetes was analysed for the candidate proteins using liquid chromatography/tandem mass spectrometry (LC/MS/MS) by high resolution multiple reaction monitoring (MRM- HR). The genetic aspects of the thesis utilised DNA extracted from the whole blood of patients from the DMP. Cases had STDR, and control participants had either no evidence of DR or only mild DR. Twenty-four single nucleotide polymorphisms (SNPs) were selected from previous genome-wide association studies (GWAS), eight VEGF and three EPO SNPs were selected from previous DR association studies, and rs4636297 has been implicated in-vivo to be associated with microvascular dysfunction. Samples were genotyped using the Sequenom iPLEX Gold chemistry on an Autoflex mass spectrometer. Association was tested using logistic regression analysis, adjusting for potential confounding variables. RESULTS: The most significant STDR protein biomarker candidates (p<0.0003) were; APOB100 (p<1.0 x 10-27), fibronectin (p<1.0 x 10-27), ceruloplasmin (p=5.34 x 10-12), gelsolin (p=9.51 x 10-12), C5 (p=4.71 x 10-8), ZAG (p=5.35 x 10-7), CD5L (p=6.57 x 10-6), protein AMBP (p=2.04 x 10-6) and ITIH1 (p=1.53 x 104) when compared to those without DR. These candidates were unable to be replicated by MRM-HR for STDR compared to without DR. Two loci from the GWAS replication were nominally associated with STDR: rs1073203 (OR=0.31, 95% CI 0.13-0.74, p=0.01 at the G allele and OR=0.24, 95% CI 0.09-0.63, p=0.003 in a dominate genotypic model), and rs4838605 (OR=1.64, 95% CI 1.02-2.65, p=0.04 in a co-dominant genotypic model). No statistically significant associations were found for VEGF and EPO polymorphisms with STDR to control participants. The miR-126-24A SNP variant was significantly associated with STDR: rs4636297 (OR=2.02, 95% CI 1.22-3.35, p=0.006 at the A allele after co-variable adjustment, and OR=2.00, 95% CI 1.19-3.38, p=0.009 in a co-dominant genotypic model). CONCLUSION: This thesis has shortlisted eight novel proteomic biomarkers, as well as one novel and two previously investigated genetic markers associated with STDR. Once replicated, we propose these biomarkers could help appropriate risk stratification of DR which would, in time allow for early diagnosis and rationalise screening programs.