Introduction There has recently been increasing interest in early life influences on the pathogenesis of cardiovascular disease. An excess of ischaemic heart disease has been observed amongst families with increased rates of stillbirth and infant mortality [1]. Recent regional variations in cardio- vascular mortality are correlated closely to regional differ- ences in infant mortality rates earlier this century [2,3]. Relatively low birth weights, even within the normal range, are associated with increased cardiovascular mor- tality [4] and with increased prevalence of hypertension [5,6], impaired glucose metabolism [7] and adverse lipid profiles [7]. The mechanisms whereby events early in life may ‘programme’ cardiovascular risk in adulthood remain unknown, but have been postulated to relate to alterations of growth during critical developmental stages [8]. Abnormalities of cardiovascular structure are frequently seen in pathological conditions associated with increased cardiovascular risk. These include hypertension [9], for which evidence suggests that abnormalities of the micro- vasculature may precede the onset of high blood pressure [10]. A recent quantitative method of characterizing the vascular networks exploits optimality principles [11] and allows the assessment of subtle changes in the retinal vasculature. A vascular network can be defined in terms of bifurcation angles and junction exponents (a measure of the relative diameters of parent and daughter branch 1111 2 3 4 5 6 7 8 9 1011 1 2 3 4 5 6 7 8 9 2011 1 2 3 4 5 6 7 8 9 3011 1 2 3 4 5 6 7 8 9 4011 1 2 3 4 5 6 7 8 9 5011 1 2 3 4 5 6 7111 Original article 1449 Retinal vascular network architecture in low-birth-weight men Neil Chapman, Anthoulla Mohamudally, Alessia Cerutti, Alice Stanton, Avan Aihie Sayer*, Cyrus Cooper*, David Barker*, Abdul Rauf † , Jennifer Evans † , Richard Wormald † , Peter Sever, Alun Hughes and Simon Thom Background Low birth weight is associated with hypertension and increased cardiovascular mortality, but the mechanism of this association is not known. Hypertension is accompanied by abnormalities of the microvasculature including rarefaction. Objective To test the hypothesis that low birth weight is associated with an alteration in microvascular architecture. Design A stratified random sample of 100 men aged 64–74 years was selected from a cohort of men whose birth weights were known. They were of relatively high or low birth weight (‘high’ > 3700 g, ‘low’ < 3200 g) and high or low systolic blood pressure (high > 160 mmHg, low < 140 mmHg). Methods Retinal arteriolar geometry was defined in terms of arteriolar bifurcation angles and junction exponents (a measure of the relative diameters of parent and daughter vessels), measured from photographic diapositives using operator-directed image analysis. Results Members of low-birth-weight groups had significantly narrower bifurcation angles than did members of high-birth-weight groups (74 ± 1° versus 78 ± 1°, P = 0.017 by analysis of variance). There was no significant difference between angles in members of groups with high and low blood pressures. Neither birth weight nor blood pressure grouping affected junction exponents. Conclusions Narrower bifurcation angles are associated with increased circulatory energy costs and may be related to a lower than normal microvascular density. Our finding of differences in retinal microvascular architecture might reflect a persistent alteration in vascular architecture as a result of an impairment of foetal development and could provide a mechanistic link between low birth weight and subsequently increased cardiovascular risk. Journal of Hypertension 1997, 15:1449–1453 Keywords: human, cardiovascular diseases, birth weight, retinal vessels, vascular networks, risk assessment From the Department of Clinical Pharmacology, Imperial College School of Medicine at St Mary’s, and the Peart-Rose Clinic, St Mary’s Hospital, London, the *MRC Environmental Epidemiology Unit, University of Southampton, and the † Department of Ophthalmic Epidemiology, Moorfields Hospital, London, UK. Sponsorship: N.C. was supported by British Heart Foundation grant PG/94046. Requests for reprints to Dr Neil Chapman, Department of Clinical Pharmacology, Imperial College of Medicine at St Mary’s, South Wharf Road, London W2 1NY, UK. Received 17 April 1997 Revised 2nd September 1997 Accepted 16 September 1997 © Rapid Science Publishers ISSN 0263-6352