Table 2.

Definitions used for carotid disease and diabetic retinopathy

First author	Diabetic retinopathy variable/s	Diabetic retinopathy assessment	Carotid disease variable/s	Carotid disease assessment
Cardoso [23]	New or worsening DR	DR graded as none, mild NPDR, moderate NPDR, severe NPDR or PDR by ophthalmologist at annual review. Worsening defined as worse by at least two grades (e.g. from mild to severe NPDR)	IMT of ICA, BIF and CCA Plaque score Number of plaques	Mean of 3 IMT measurements used, IMT measured according to Mannheim consensus Extracranial carotid artery plaque score assigned for each segment of ICA, CCA and BIF and highest grade assigned; 0: no plaque, 1: one small plaque, stenosis < 30%, 2: one medium plaque, 30–49% stenosis or multiple small plaques, 3: one large plaque, 50–99% stenosis or multiple plaques with at least one medium plaque, 4: 100% occlusion Single vascular radiologist performed ultrasound, good intra-observer test–retest reliability
Hjelmgren [32]	Any	Medical records from all eye clinics in the area	Stenosis > 50%	Greater stenosis from left or right artery used
Ichinohasama [8]	Mild NPDR versus no DR	Right eye assessed by ophthalmologist, according to ETDRS protocol	CCA IMT	Maximum measurement of right CCA IMT
Carbonell [7]	None, mild or advanced (moderate NPDR or worse)	Examination by ophthalmologist, according to ETDRS protocol	Plaque	Plaque was defined according to the Mannheim consensus Single sonographer at each study site performed ultrasound
Liu [34]	Any	Retinal images graded by ophthalmologist according to International Clinical Diabetic Retinopathy Disease Severity Scale	CCA IMT > 1 mm Plaque Subclinical atherosclerosis	Maximum CCA IMT value of left or right artery used, IMT was measured 1.5 cm proximal to the bifurcation, elevated CCA IMT defined as > 1 mm Plaque was classified as focal increase in thickness more than 0.5 mm or 50% of surrounding IMT Subclinical atherosclerosis was defined as CCA IMT > 1 mm and/or presence of carotid plaque Single sonographer performed ultrasound
Alonso [28]	Any and mild NPDR, moderate NPDR or severe NPDR or PDR	Multi-field stereoscopic retinal images and ophthalmologist examination	IMT of ICA, BIF and CCA Plaque	Semiautomatic software provided data for mean and mean-maximum IMT from segments of ICA, BIF and CCA. Values from the left and right arteries were averaged Plaque was defined according to the Mannheim consensus Single sonographer performed ultrasound
Jung [27]	Any	Examination by ophthalmologist	CCA IMT ≥ 1 mm, Plaques > 2	CCA IMT measured 1 cm proximal to bulb on left and right, mean of these were used Protrusions > 100% defined as plaque. Plaques were categorised into n ≤ 2 and n > 2 plaques
Cardoso [29]	Any	Examination by ophthalmologist	IMT of CCA, BIF, ICA Plaque score ≤ 2 or > 2	Mean of 3 IMT measurements used. IMT measured according to Mannheim consensus Extracranial carotid artery plaque score assigned for each segment of ICA, CCA and BIF and highest grade assigned; 0: no plaque, 1: one small plaque, stenosis < 30%, 2: one medium, plaque 30–49% stenosis or multiple small plaques, 3: one large plaque, 50–99% stenosis or multiple plaques with at least one medium plaque, 4: 100% occlusion Single vascular radiologist performed ultrasound, good intra-observer test–retest reliability
Yun [10]	Any	Retinal images graded according to ETDRS protocol	CCA IMT Plaque	CCA IMT measured by software at thickest point ~ 1 cm from bulb, analysed in tertiles Plaque defined as protrusions into lumen that were 100% thicker than surrounding area Physicians performed ultrasound
Son [10]	Any	Two-field retinal images and ophthalmologist examination	CCA IMT ≤ 0.9 mm and no plaque compared to CCA IMT > 0.9 mm ± carotid plaque	IMT measured bilaterally 5–10 mm proximal to bulb, 3 measurements done at site of greatest thickness and 10 mm proximal and distal to this point, highest mean CCA IMT used Plaque was a focal increase of ≥ 0.5 mm or ≥ 50% of surrounding IMT Single sonographer performed ultrasound
Araszkiewicz [30]	Any	Two-field retinal images and ophthalmologist examination according to American Academy of Ophthalmology	CCA IMT	Right CCA IMT measured and automatically calculated with software program—Carotid Analyzer for Research (CAD 5)
Lacroix [33]	Any	Examination by ophthalmologist	no atherosclerotic lesion or stenosis < 60% or stenosis ≥ 60%	Stenosis was considered ≥ 60% when the maximal velocity within the lesion was > 2.6 m/s and the end-diastolic velocity > 0.7 m/s Performed by experienced vascular physicians
Distiller [31]	Any	Retinal images assessed	CCA IMT Plaque IMT risk (low, medium or high)	CCA IMT measured > 1 cm proximal to flow divider, mean of left and right used Plaque defined as localised thickening of wall of ≥ 1.5 mm IMT risk: low < 0.6 mm, medium 0.6–0.8, high > 0.8 and/or plaque Two sonographers, intra-observer & inter-observer variability 3.1% and 3.9%
Rema [9]	Any	Four-field retinal images graded according to ETDRS protocol by two graders, a third grader made final decision if discrepancy	IMT of CCA, BIF, ICA	Mean of six IMT measurements of right ICA, CCA and BIF All scans were quality controlled by a central laboratory in Canada

DR diabetic retinopathy, NPDR non-proliferative diabetic retinopathy, PDR proliferative diabetic retinopathy, CSMO clinically significant macular oedema, VTDR vision-threatening diabetic retinopathy, ETDRS Early Treatment Diabetic Retinopathy Study, CCA common carotid artery, ICA internal carotid artery, IMT intima-media thickness