Elsevier

The Lancet

Volume 358, Issue 9288, 6 October 2001, Pages 1134-1140
The Lancet

Articles
Retinal microvascular abnormalities and incident stroke: the Atherosclerosis Risk in Communities Study

https://doi.org/10.1016/S0140-6736(01)06253-5Get rights and content

Summary

Background

Retinal microvascular abnormalities reflect damage from hypertension and other vascular processes. We examined the relation of such abnormalities to incident stroke.

Methods

A cohort of 10 358 men and women (aged 51 to 72 years) living in four US communities underwent retinal photography and standard grading for retinal microvascular abnormalities. The calibres of all retinal arterioles and venules were measured after digital conversion of the photographs, and a summary arteriole-to-venule ratio (AVR) was calculated as an index of arteriolar narrowing (smaller AVR indicates greater narrowing). Cases of incident stroke admitted to hospital were identified and validated by case record reviews.

Findings

Over an average of 3.5 years, 110 participants had incident strokes. After adjustment for age, sex, race, 6-year mean arterial blood pressure, diabetes, and other stroke risk factors, most retinal microvascular characteristics were predictive of incident stroke, with adjusted relative risks of 2.58 (1.59-4.20) for any retinopathy, 3.11 (1.71-5.65) for microaneurysms, 3.08 (1.42-6.68) for soft exudates, 2.55 (1.27-5.14) for blot haemorrhages, 2.26 (1.00-5.12) for flame-shaped haemorrhages, and 1.60 (1.03-2.47) for arteriovenous nicking. The relative risk of stroke increased with decreasing AVR (p=0.03). The associations were similar for ischaemic strokes specifically, and for strokes in individuals with hypertension, either with or without diabetes.

Interpretation

Retinal microvascular abnormalities are related to incident stroke. The findings support microvascular role in the pathogenesis of stroke. They suggest that retinal photography may be useful for cerebrovascular-risk stratification in appropriate populations.

Introduction

Retinal microvascular abnormalities (such as microaneurysms and arteriovenous nicking) reflect arteriolar damage from hypertension and other processes. They have been suggested as markers for cerebral microvascular diseases, because the retinal and cerebral arterioles share common anatomy and physiology.1, 2, 3 Previous epidemiological studies that have investigated the association between retinal microvascular abnormalities and stroke have shown inconsistent results.4, 5, 6, 7, 8 This inconsistency may result partly from use of clinical ophthalmoscopy, a subjective and unreliable technique, to detect retinal lesions.9, 10, 11 Moreover, some of these data may not apply to contemporary populations, with lower blood pressures and infrequent signs of advanced retinopathy.12

Retinal photography was done for all participants at the third examination (1993-95) of the Atherosclerosis Risk in Communities (ARIC) study, a large populationbased cardiovascular study in middle-aged people. These photographs were graded for the presence of retinal microvascular characteristics on the basis of a standard protocol.13, 14, 15 The severity of retinal arteriolar narrowing was further quantified by measurement of individual arteriolar and venular calibres on highresolution digitised photographs.13 We have previously shown that this grading system is reliable13 and that retinal microvascular abnormalities are related to current blood pressure, to past blood pressure independently of current blood pressure,14 and to various markers of inflammation and endothelial dysfunction.15

In this study, we describe the relations between retinal microvascular abnormalities and incident stroke in the ARIC population.

Section snippets

Methods

Study population The ARIC study included a cohort of 15 792 women and men selected as probability samples of 45-64-year-old residents of four US communities: Forsyth County, NC; Jackson, MS (black people only); suburbs of Minneapolis, MN; and Washington County, MD.16 The initial participation rate was 46% in Jackson and about 65% in the other communities. Of the people examined at baseline, 14 346 (93% of survivors) returned for a second examination 3 years later and 12 887 (86% of survivors)

Procedures

Retinal photography and the grading procedure have been described in detail elsewhere.13, 14, 15 Briefly, after 5 min of dark adaptation, a 45° retinal photograph was taken of one randomly selected eye, centred on the region of the optic disc and the macula, with an autofocus camera. The photographic slides were assessed by trained graders, unaware of the participant's identity. The following lesions were graded: arteriovenous nicking, focal arteriolar narrowing, blot haemorrhages, flameshaped

Statistical analysis

All statistical procedures used SAS software (version 8.0). Follow-up time was defined as the number of days from the third examination visit to the date of hospital admission for the first stroke, death, last contact, or Dec 31, 1997, whichever occurred first.

We calculated Kaplan-Meier cumulative survival comparing presence versus absence of a focal retinal lesion, or between quintiles of AVR, and presented the result as 100x(1-Kaplan-Meier estimators). Relative risks for stroke were derived

Results

Comparisons of baseline characteristics between participants included (n=10 358) and excluded (n=5434) from the study are given in table 1. In general, those included were younger and had more years of education, the proportion who were white was higher, the proportions with hypertension, diabetes, and coronary artery disease were lower, and the mean values of blood pressure, body-mass index, and total cholesterol, triglyceride, and glucose concentrations were lower.

Among the participants

Discussion

Stroke affects an estimated 600 000 people per year in the USA and is the leading cause of severe neurological disability.21 Despite extensive research, the contribution of microvascular disease to the pathogenesis and evolution of stroke is not clear.23, 24 The retinal arterioles offer a unique opportunity to investigate the role of the microcirculation in cerebrovascular diseases, because they are accessible to direct non-invasive assessment, and they share anatomical and physiological

References (28)

  • SvardsuddK et al.

    Hypertensive eye ground changes: prevalence, relation to blood pressure and prognostic importance

    Acta Med Scand

    (1978)
  • OkadaH et al.

    A prospective study of cerebrovascular disease in Japanese rural communities, Akabane and Asahi, part 1: evaluation of risk factors in the occurrence of cerebral hemorrhage and thrombosis

    Stroke

    (1976)
  • TanakaH et al.

    Epidemiologic studies of stroke in Shibata, a Japanese provincial city: preliminary report on risk factors for cerebral infarction

    Stroke

    (1985)
  • NakayamaT et al.

    A 15·5-year follow-up study of stroke in a Japanese provincial city: the Shibata Study

    Stroke

    (1997)
  • Cited by (0)

    View full text