Effect of birth parameters on retinal vascular caliber: the twins eye study in Tasmania

Recent studies reported an association between smaller birth size and narrower retinal vascular caliber, but it remains unclear whether this association is attributed to confounding by shared environment or genetic factors. At a mean age of 9.3 years, 266 twins (49 monozygotic and 84 dizygotic pairs) in the Twins Eye Study in Tasmania underwent an ophthalmic examination including retinal photography. Retinal vascular caliber was measured using a validated protocol. The majority of these twins were also in the Tasmanian Infant Health Study, which prospectively collected data on birth parameters and antenatal maternal factors. We conducted the main analysis using linear mixed models, accounting for birth set clustering. Both the within-pair (-9.73; 95% CI: -14.68 to -4.77 mum per 5-cm decrease in birth length) and between-pair associations (-7.15; 95% CI: -11.54 to -3.01) with retinal arteriolar caliber were significant and of similar magnitude (difference in effect, P=0.61), after adjusting for age, sex, maternal smoking, mean arterial blood pressure, and other confounders. These associations remained within dizygotic and monozygotic pairs. Analyses of head circumference and retinal arteriolar caliber were similar to those of birth length (within-pair regression coefficient: -2.41; 95% CI: -5.09 to 0.28; between-pair regression coefficient: -2.60; 95% CI: -5.00 to -0.19). For birth weight, only a between-pair association was evident (-7.28; 95% CI: -13.07 to -1.48). This study demonstrates a consistent association between smaller birth size and narrower retinal arterioles in twins. The independent effect of shorter birth length on retinal arteriolar caliber supports a role for twin-specific supply line factors affecting fetal growth on vascular structure.