Table 2. Relation of retinal vascular calibers, presence of retinopathy and presence of microalbuminuria on risk of future cardiovascular disease event.

	No. at risk	Incident cases, %	HR (95% CI)*	HR (95% CI)†
Retinal arteriolar caliber (μm)
Quartile 4 (151.08 − 206.31)	876	28 (3.20)	Referent	Referent
Quartile 3 (141.72 − 151.06)	873	18 (2.06)	0.63 (0.35, 1.15)	0.75 (0.41, 1.38)
Quartile 2 (132.08 − 141.71)	873	28 (3.21)	0.82 (0.48, 1.38)	1.03 (0.59, 1.80)
Quartile 1 (71.68 − 32.07)	874	52 (5.95)	1.03 (0.71, 1.80)	1.50 (0.87, 2.58)
P for trend			0.328	0.071
per SD decrease (14.60)	3496	126 (3.60)	1.05 (0.89, 1.24)	1.20 (0.98, 1.47)
Retinal venular caliber (μm)
Quartile 1 (104.03−206.34)	874	24 (2.75)	Referent	Referent
Quartile 2 (206.37 − 220.11)	874	27 (3.09)	1.33 (0.77, 2.31)	1.23 (0.69, 2.19)
Quartile 3 (220.12 − 233.33)	874	33 (3.78)	1.05 (1.01, 3.02)	1.64 (0.93, 2.89)
Quartile 4 (233.34 − 296.67)	874	42 (4.81)	2.25 (1.35, 3.76)	2.07 (1.15, 3.73)
P for trend			0.001	0.008
per SD increase (20.70)	3496	126 (3.60)	1.32 (1.13, 1.55)	1.33 (1.09, 1.62)
Presence of Retinopathy
No	3189	101 (3.17)	Referent	Referent
Yes	307	25 (8.14)	2.20 (1.42, 3.41)	2.05 (1.30, 3.22)
	No. at risk	Incident cases, %	HR (95% CI)*	HR (95% CI)‡
Microalbuminuria
No	2,991	81 (2.71)	Referent	Referent
Yes	505	45 (8.91)	2.27 (1.57, 3.32)	1.68 (1.13, 2.50)

^*Adjusted for age, gender and ethnicity.

^†Further adjusted for current smoking, diabetes, total cholesterol, HDL cholesterol, systolic blood pressure, anti-hypertensive medication, eGFR, retinal arteriolar caliber (when the main exposure is retinal venular caliber and vice versa) and retinopathy.

^‡Further adjusted for current smoking, diabetes, total cholesterol, HDL cholesterol, systolic blood pressure, anti-hypertensive medication, eGFR, retinal venular caliber and retinopathy. HR: Hazard Ratio.