Ophthalmology (Eye & Ear Hospital)

PURPOSE Luminance flicker-induced retinal vasodilation is reduced in people with type 1 diabetes compared with non-diabetic controls. However, the mechanisms of this response and causes for its reduction are not well understood. I therefore investigated potential acute causes of reduced luminance flicker-induced retinal vasodilation in people with and without type 1 diabetes. My doctoral project had four specific aims: 1) To investigate the effect of repeated testing on luminance flicker-induced retinal vasodilation in healthy people; 2) To study the effect of ambient lighting on luminance flicker-induced retinal vasodilation in healthy people; 3) To determine the importance of epoxyeicosatrienoic acids (EETs) and prostaglandins (PGs) in luminance flicker-induced retinal vasodilation in healthy people; and 4) To study the effect of glucose levels and antioxidant treatment on retinal flicker responses, ganglion cell function and systemic arterial elasticity in people with type 1 diabetes. METHODS My doctoral project involved a total of 72 participants (60 non-diabetic and 12 with type 1 diabetes). I first assessed the impact of repeated testing with five or 30 minutes between tests on luminance flicker-induced retinal vasodilation in 20 healthy people. Next, I used a balanced crossover study investigated the effect of reduced and normal ambient lighting on these responses in another 20 healthy people. In my third study, I used oral fluconazole 400 mg and dispersible aspirin 600 mg to investigate the role of EETs and PGs, respectively, in healthy luminance flicker-induced retinal vasodilation. Twelve healthy people participated in a balanced crossover study with three treatments (no drug, fluconazole and aspirin). In addition, six healthy people, each for fluconazole and aspirin, were followed at 30-minute intervals for two hours after drug ingestions. My final study involved 12 otherwise healthy people with type 1 diabetes. I investigated luminance flicker-induced retinal vasodilation, retinal ganglion cell function and systemic arterial elasticity at baseline, euglycaemia (plasma glucose = 6 mmol/l) and hyperglycaemia (plasma glucose = 15 mmol/l). Participants were seen twice: once with vitamin C (2 g intravenous) and once with placebo immediately prior to the initiation of hyperglycaemia. Retinal vasodilation in response to 20 seconds of 12.5 Hz luminance flicker with green light was assessed with the Dynamic Vessel Analyzer (DVA) after pupil dilation with tropicamide 1%. Ganglion cell activity was measured from the pattern electroretinogram (PERG) transient (1 Hz stimulus) and steady-state (8.33 Hz) responses to black and white check reversals without pupil dilation. Systemic arterial elasticity was also assessed by radial artery pulsewave tonometry with the HD/Pulsewave CR-2000. Participant characteristics were compared by analysis of variance (ANOVA) for continuous variables or Fisher’s exact test for categorical variables. Within-subject changes in group means were assessed by ANOVA. Significant differences identified by ANOVA were explored by post-hoc Student’s t-tests with corrections for multiple comparisons where necessary. Within-subjects Pearson’s correlation coefficients were assessed in my final study for luminance flicker-induced retinal vasodilation, ganglion cell function and systemic arterial elasticity. Two-tailed P < 0.05 was considered statistically significant. All statistical analyses were performed in STATA version 12.1. RESULTS Aim 1 Twenty healthy people aged (mean [standard deviation [SD]]) 33.1 (5.7) years were studied. Maximum luminance flicker-induced arteriolar dilations were 3.2 (2.1) % initially, 2.4 (1.6) % after five minutes of rest and 3.4 (2.1) % after 30 minutes of rest (P < 0.001 by ANOVA). Compared with the first test, arteriolar dilations were reduced after five minutes (P = 0.02 after Bonferroni’s adjustment), but not after 30 minutes of rest (P > 0.05). Maximum venular dilations were 4.3 (1.3) % initially, 3.8 (1.6) % after five minutes of rest and 4.4 (1.7) % after 30 minutes of rest (P > 0.05 by ANOVA). Pre-flicker vessel diameters were unchanged between tests (P > 0.05 for arterioles and venules). Aim 2 Twenty healthy people aged 32.8 (6.5) years were studied. Maximum luminance flicker-induced retinal arteriolar dilations were 4.8 (2.3) % and 4.1 (1.9) % under reduced and normal ambient lighting, respectively (P = 0.02 by ANOVA). Corresponding maximum venular dilations were 5.4 (1.8) % and 5.1 (2.1) % under reduced and normal ambient lighting, respectively (P > 0.05). Pre-flicker arteriole and venules diameters were unchanged between tests (P > 0.05 for both). Aim 3 Twenty healthy people aged 25.6 (4.6) were studied. In 12 crossover study participants, maximum luminance flicker-induced retinal arteriolar and venular dilations without drug administration were 4.4 (2.0) % and 4.6 (1.7) %, respectively. Neither fluconazole nor aspirin affected these responses (P > 0.05 for both by ANOVA). Pre-flicker arteriole and venule diameters without drug administration were 120 (11) measurement units (MU) and 146 (17) MU, respectively. Fluconazole reduced pre-flicker venule diameters by (mean ± 95% confidence interval [CI]) 5 ± 4 MU (P = 0.02 after Dunnett’s adjustment). In six participants, fluconazole did not affect arteriolar or venular dilations or pre-flicker arteriole diameters (P > 0.05 for all) but did reduce pre-flicker venule diameters over two hours (P < 0.001 by ANOVA). Aspirin did not affect arteriolar or venular dilations or pre-flicker diameters in six participants over two hours (P > 0.05 for all). Aim 4 Twelve otherwise healthy people with type 1 diabetes aged (median [interquartile range [IQR]]) 24 (20.5 – 30) years with 10.5 (3.8 – 17) years of diabetes were studied. Plasma glucose levels decreased from (mean [SD]) 8.4 (3.6) mmol/l at baseline to 6.1 (0.8) mmol/l at euglycaemia (P < 0.01). Plasma glucose levels then increased during hyperglycaemia to 15.0 (1.9) mmol/l with vitamin C and 14.4 (0.8) mmol/l with placebo (both P < 0.001 vs. euglycaemia). Insulin levels increased from 65 (24) pmol/l at baseline to 334 (63) pmol/l at euglycaemia (P < 0.001) and were stable thereafter. Euglycaemic clamp increased retinal venule maximum dilation by 1.7 (2.0) % and area under the curve (AUC) during flicker by 28.7 (28.9) % x sec. compared to baseline (both P < 0.01). No change in flicker responses was observed between euglycaemia and either hyperglycaemia arm or between vitamin C and placebo hyperglycaemia arms. Ganglion cell function and systemic arterial elasticity were unaffected across glucose conditions (P > 0.05 for all). The one exception was the ratio of the second harmonic frequency (2F) amplitude with 0.8° check sizes to the amplitude with 7° check sizes from the PERG steady-state response. The ratio increased from 1.1 (0.3) at baseline to 1.4 (0.6) after hyperglycaemia with vitamin C (P < 0.05). In within-subjects correlation analyses, the arteriole AUC during flicker was positively correlated with plasma glucose levels (r = 0.27, P = 0.04), although this correlation became non-significant when data from hyperglycaemia with vitamin C were excluded. Arteriole and venule maximum dilations and AUC during flicker were positively correlated with free insulin levels, although only venule dilation responses remained significant after data from hyperglycaemia with vitamin C were excluded. Large artery elasticity (LAE) was positively correlated with arteriole maximum dilations (r = 0.26, P = 0.04) and AUC during flicker (r = 0.26, P = 0.04) and these correlations remained significant after the exclusion of data obtained after vitamin C treatment. Both LAE (r = -0.32, P = 0.01) and small artery elasticity (SAE) (r = -0.35, P = 0.01) were negatively correlated with systolic blood pressure. These correlations were unchanged by the exclusion of vitamin C data. In PERG tests, the N95 amplitude was positively correlated with diastolic blood pressure, the 0.8° 2F amplitude was negatively correlated with systolic blood pressure and the 7° 2F amplitude was negatively correlated with insulin levels. After the exclusion of vitamin C data, the correlation between the 0.8° 2F amplitude and systolic blood pressure was not statistically significant. CONCLUSIONS Several factors may impact on luminance flicker-induced retinal vasodilation in humans. Arteriolar responses, in particular, are reduced with repeated stimulation and higher ambient lighting. This may reflect light adaptation in photoreceptors and lower modulation depths during luminance flicker, respectively, with secondary reductions in ganglion cell activation. My data do not support a major role for EETs and PGs in luminance flicker-induced retinal vasodilation. Nitric oxide remains the most likely signalling molecule and dysfunction in its signalling pathways may contribute to reduced responses in type 1 diabetes. Increased venule dilations during euglycaemic clamp suggest that glucose normalisation with insulin improves retinal microvascular function. Furthermore, antioxidants such as vitamin C may improve retinal neurovascular function in adults with type 1 diabetes. This finding indirectly implicates oxidative stress as a contributor to impaired retinal neurovascular function in type 1 diabetes. My doctoral project highlights several important factors that may reduce luminance flicker-induced retinal vasodilation and may lead to its use as a clinical marker of neurovascular function in retinal diseases.

There is controversy over observations that myopia is protective of diabetic retinopathy (DR) due to the uncertainty over the contributory roles of ocular biometric parameters (i.e. axial length [AL], corneal curvature [CC], and anterior chamber depth [ACD]) all of which are closely associated with myopia. Moreover, theories on the mechanisms underpinning the supposed protective myopia-DR relationship are educated guesses inferred from observations of reduced retinal blood flow and/or oxygen (O2) consumption in myopic eyes. Therefore, this thesis has three main aims. The first is to investigate the associations of myopia and myopia-related ocular biometric parameters with DR (and diabetic macular edema [DME]). The second and third aims are to assess the respective roles of a reduction in retinal capillary blood flow (RCF) and O2 consumption as potential mechanisms underlying this protective myopia-DR relationship. Aim one included 609 patients with diabetes aged 18+ years. After multivariable adjustments, eyes with longer AL were less likely to have mild (odds ratio [OR]: 0.58, 95% confidence interval [CI]: 0.41- 0.83 per mm increase), moderate (OR: 0.73, 95% CI: 0.60-0.88), and severe DR (OR: 0.67, 95% CI: 0.53-0.85), and also had a lower risk of mild (OR: 0.70, 95% CI: 0.56- 0.86); and moderate DME (OR: 0.72, 95% CI: 0.56- 0.93). No associations were found for refractive error, CC or ACD with DR, suggesting that the myopia-DR association is likely due to a longer AL in myopic eyes. For aim two, I utilized 85 persons with diabetes aged 18+ years, independent of the sample used in aim one. Evaluating the relationship between AL, RCF and DR, I found no association between AL and RCF (per mm increase in AL, regression coefficient [β] -1.80, 95% CI: -13.50-9.50) or between RCF and the risk of DR (per unit increase in RCF, OR 1.00, 95% CI 0.99-1.00) after multivariable adjustments. These findings suggest that diminished RCF may not be a major mechanism underlying the protective association between axial elongation and DR. In aim three, I utilized the arterio-venous (A-V) difference in O2 saturation (SO2), as a surrogate marker of relative O2 consumption. A reliability and reproducibility study in 20 healthy individuals aged 18+ years found extremely high intra-class correlation coefficient (ICC) values for intra-observer reliability (0.99 for both arteriolar and venular SO2), inter-observer reliability (ICC of 0.94 for arteriolar SO2 and 0.96 for venular SO2) and intra-subject reproducibility (ICC of 0.98 for both arteriolar and venular SO2) for the retinal oximeter utilized in this doctoral thesis. Finally I explored the role of O2 consumption as one of the mechanisms underpinning the protective association between AL and DR. Due to the potential confounding effects of diabetes on retinal function, the relationship between AL, retinal function (assessed using multifocal electroretinogram [mfERG] P1 amplitude) and the A-V difference was conducted in 50 healthy subjects aged 18+ years. Path analysis models including AL (study factor), retinal function (intermediate variable) and A-V difference (outcome variable) showed that AL had little direct association with A-V difference (βp = -0.002), while the indirect effect of AL on A-V difference via changes in retinal function were substantial (βp = -0.51). These findings suggest that longer eyes have decreased retinal function and O2 consumption, and thus are relatively less hypoxic in the presence of diabetes, which may partly explain the reduced risk of DR in these eyes. In conclusion, I have demonstrated that a longer AL is the main factor contributing to the protective association of myopia with risk of DR. Furthermore, I have established that the protective myopia-DR relationship may be partially due to the alleviation of hypoxia in diabetes, due to a reduction in retinal function as the eye elongates.

Ophthalmology (Eye & Ear Hospital) - Theses

Permanent URI for this collection

Filters

Date

Author

Subject

Type

Settings

Sort By

Results per page

Statistics

Citations

Search Results