Published in:
01-10-2017 | Article
Innovative technology shows impact of glycaemic control on peripheral retinal vessels in adolescents with type 1 diabetes
Published in: Diabetologia | Issue 10/2017
Login to get accessAbstract
Aims/hypothesis
Retinal imaging enables non-invasive microvasculature assessment; however, only central retinal vessels have been studied in type 1 diabetes. Peripheral smaller vessels have a major haemodynamic role and may differ from central vessels in their response to the diabetic milieu. We hypothesise that diabetes has a greater impact on peripheral retinal vessels vs central vessels.
Methods
Retinal photographs from adolescents (n = 736; age 12–20 years) with type 1 diabetes were graded (Singapore I Vessel Assessment) with vessel calibres measured in the ‘central zone’ as central retinal arteriolar and venular equivalents (CRAE and CRVE, respectively) and the ‘extended zone’ as mean width of arterioles and venules (MWa and MWv, respectively). Multivariable linear regression was used to explore associations between vessel calibres and HbA1c, diabetes duration, sex and BP.
Results
Mean ± SD age was 14.1 ± 1.5 years, HbA1c was 8.5 ± 1.3% (69.4 ± 14.1 mmol/mol) and median diabetes duration was 4.9 years (interquartile range 3.1–7.6 years). Wider MWa was associated with HbA1c (β 0.01 [95% CI 0.004, 0.03]), longer diabetes duration (0.07 [0.02, 0.13]) and higher systolic BP (0.04 [0.02, 0.05]). MWv was associated with HbA1c (0.02 [0.009, 0.03]) and higher systolic BP (0.04 [0.03, 0.06]). CRAE was associated with longer diabetes duration (0.93 [0.58, 1.28]) and higher systolic BP (−0.28 [−0.37, −0.19]). CRVE was associated with longer diabetes duration (0.91 [0.42, 1.41]) and higher systolic BP (−0.20 [−0.33, −0.07]). Girls had wider vessels (for all four calibre measurements).
Conclusions/interpretation
In adolescents with type 1 diabetes, higher HbA1c is associated with adverse changes to peripheral smaller retinal vessels but not central vessels. The predictive value of retinal vascular imaging should be evaluated using longitudinal data.