Background
The deleterious effects of hyperglycaemia are classically separated into microvascular (retinopathy, diabetic nephropathy and neuropathy) and macrovascular complications (coronary artery disease, peripheral arterial disease and cerebrovascular disease).
Traditionally, the micro- and macrovascular complications of diabetes have been viewed, studied, and treated as distinct and independent disorders. However, accumulating data from epidemiological and pathophysiological studies suggest that these vascular complications may share common pathophysiological mechanisms beyond those related to traditional cardiovascular (CV) risk factors. Data obtained from epidemiological studies have clearly demonstrated that diabetic retinopathy (DR), a common chronic microvascular complication of diabetes, is associated with macrovascular disease [
1] as well as with increased CV morbidity and mortality in patients with type 2 diabetes mellitus (T2D) [
2]. In these patients, the presence of DR has been described as an independent risk factor for incident coronary heart disease [
3,
4] and ischaemic stroke [
5]. Among the possible mechanisms described to explain the interconnection between diabetic micro- and macroangiopathy are metabolic disturbances [
6]. Other authors have suggested that microangiopathy of the vasa vasorum (VV), the plexus of microvessels that partially provides oxygen and nutrients to the walls of large arteries, may be involved in diabetic atherosclerosis [
7,
8]. Our group recently reported that T2D patients with DR had increased angiogenesis of the VV of the common carotid artery (CCA) [
9]. These findings were recently confirmed by Sampson
et al. in a larger study of T2D patients [
10].
When a clinical CV event occurs, atherosclerotic disease is difficult to reverse. In these cases, ultrasonography of the carotid arteries is frequently used to detect early signs of atherosclerosis, i.e., increased thickness of the arterial wall or the occurrence of plaques. Ample data suggest that plaque and carotid intima-media thickness (cIMT) are associated with prevalent and incident coronary heart disease (CHD) and stroke, with the presence of plaque generally having a stronger association with CVD compared to cIMT alone [
11]. Thus, it has recently been demonstrated that ultrasound assessment of a carotid plaque and its total volume or total area progression is a stronger predictor of future CV events than cIMT measurement [
12,
13]. A high prevalence (between 43% and 64%) of carotid plaques has been described in T2D patients without evidence of CV disease (CVD) [
14-
16]. Prospective studies conducted in T2D patients free of any CV event have shown that the percentage of patients with carotid artery plaques is higher in those who develop a CV event compared with those who are free from CV events [
15,
16]. Recent epidemiological studies have demonstrated that chronic kidney disease (CKD), defined as an estimated glomerular filtration rate (eGFR) <60 ml/min, is an independent risk factor for atherosclerotic disease [
17] and for the presence of carotid plaques [
18]. In T2D patients, a high prevalence of low eGFR without associated albuminuria has been reported and described to be associated with atherosclerosis of peripheral arteries, independently of albuminuria, [
19] and with cIMT in several [
20], but not all, studies [
21,
22].
To test the concept that patients with T2D and DR have an increased prevalence of subclinical atherosclerosis, as shown using carotid plaque burden and cIMT measurements, we investigated the association between these measures and retinopathy in a group of patients with T2D but without previous CVD and with normal renal function.
Discussion
The present study shows that T2D patients with DR who are free of clinical CVD and with normal renal function have increased atherosclerotic burden in their carotid arteries compared with patients without DR.
Epidemiological studies have demonstrated that DR in T2D patients is associated with a 1.7-fold increased risk of CV events, such as stroke, coronary artery disease and heart failure [
26]. Notably, this risk persists after adjusting for traditional CV risk factors, diabetes duration and diabetes control, suggesting that microvascular disease may contribute to the development of CVD in diabetes. A major finding of the present study is that DR is independently associated with the presence of carotid plaques. This result is similar to that observed in the study by Kreutzenberg
et al., in which DR was independently associated with carotid plaques, although the number of patients with retinopathy was lower in that study than in our study, and patients with previous CV events were not excluded [
27]. On the other hand, several studies have described an association between cIMT thickness, as a surrogate marker of atherosclerosis, and DR that is independent of traditional risk factors and glucose control [
28,
29]. Miyamoto
et al. also reported an association between common carotid arterial diameter, as a marker of atherosclerosis, and the presence of DR [
30]. It should be noted that in the present study, the percentage of patients with carotid plaques was not only higher in those with DR than in those without DR but that the number of carotid plaques was also elevated. To the best of our knowledge, there is no report in the literature regarding atherosclerotic plaque burden in T2D patients with or without DR. A direct association has been described between a plaque score, consisting of the sum of the maximum plaque thicknesses of different carotid territories, and the presence and extent of coronary artery disease in T2D patients [
31]. As reported in general-population studies, the risk of CV events, such as stroke and cerebral infarction, gradually increases with incremental total plaque number, irrespective of the location of the plaques [
32]. Moreover, carotid plaque measurements have been shown to be more strongly predictive of CV events than cIMT measurements. The extent of atherosclerosis in the carotid artery, measured either as total plaque area [
30] or as progression of total plaque volume [
12], has been shown to be an independent predictor of coronary events in patients with [
12] and without previous CV events [
33]. Although these studies included a small proportion of subjects with diabetes, no specific data on these subjects were provided. In the Rotterdam study, a direct relationship between the number of carotid territories with a carotid plaque and incident cases of myocardial infarction was described [
34].
Carotid plaque predominantly occurs at sites of nonlaminar turbulent flow, such as the carotid bulb and the proximal ICA, but rarely in the CCA, except in advanced atherosclerotic disease [
35]. In the present study, the percentage of plaques in the ICA was higher in T2D patients with DR than in those without DR; however, these differences were not significant. The presence of plaques in the ICA territory is now accepted to be associated with an increased risk of stroke (annual risk of 0.1% to 1.9% for an ICA stenosis <75% to 80%, and annual risk of 2.0% to 3% with greater degrees of stenosis) [
36]. Thus, considering research results suggesting that CV events increase as the burden of atherosclerosis increases, we hypothesise that T2D patients with DR and >1 carotid plaque will have a higher risk of CV events. This speculation can only be answered by conducting a prospective study assessing the future occurrence of CV events in these patients.
As stated in almost all studies on DR, the condition is closely related to duration of the disease, higher blood pressure and a higher albumin excretion rate, all of which may contribute to the increased atherosclerosis in these patients [
37]. In the present study, these factors were also more frequent in patients with DR. However, of these variables, only hypertension proved to be independently associated with carotid plaques.
It is now accepted that CKD, defined as an eGFR of <60 ml/min, is an independent risk factor for atherosclerotic diseases and is associated with the presence of a carotid plaque independent of albuminuria [
38]. Recent studies have emphasised that a large number of T2D patients with low eGFR have a normal UAE [
39,
40]. It should be noted that little information regarding renal function has been reported in those studies that analysed subclinical atherosclerosis in patients with T2D. In this respect, the study by Saif
et al., conducted inT2D patients with normal renal function, did describe a correlation between DR and cIMT [
29].
Another ultrasound measure that is used as a surrogate marker for atherosclerosis is cIMT. This measure, as described for carotid plaques, has been reported to be associated with traditional and newer risk factors for atherosclerosis. The cIMT baseline measures and its progression during follow-up are described to be strongly associated with CVD events, even after adjustment for traditional risk factors [
41]. cIMT has typically been measured in the CCA, rather than in the carotid bulb or ICA, because high-level measurement precision is easily obtained from this artery. However, it should be noted that the accuracy of cIMT as a marker of atherosclerosis has been questioned by the fact that medial hypertrophy or intimal thickening of CCA may be influenced by factors that do not necessarily reflect the atherosclerotic process, such as age and hypertension [
42]. In the present study, T2D patients with DR had a higher mean-maximum common cIMT than did those without DR, whereas no differences were observed in the bulb-cIMT or in the internal cIMT measures between the groups. However, DR was independently associated with only the mean-internal cIMT. In relation to cIMT, it is accepted that, clinically, the measurement of CCA-cIMT is easier and more reliable compared to ICA-cIMT. In contrast, ICA-cIMT appears to predict atherosclerotic cardiac events better than common cIMT [
43]. Thus, including internal cIMT and traditional CV risk factors, as well as the presence of plaque in the internal territory, improves the risk classifications of stroke and coronary heart disease in general population studies.
The association between DR and carotid ultrasound measurements has been analysed in large-population cohort studies conducted on subjects without a history of clinical CVD. The ARIC study found that the severity of retinopathy was associated with carotid artery intima-media wall thickness [
44], while the CHS [
45] and MESA [
46] studies found no association between DR and cIMT or carotid stenosis. Possible explanations for the different results obtained in these last two studies and ours may be due to the difference in ethnic backgrounds of patient groups, different criteria used for the definition of subclinical atherosclerosis, different population selection or the absence of bulb territory (a site with a high prevalence of plaques) in the ultrasound examination in these studies. In the present study, no differences in cIMT measurements or in plaque presence were observed among the groups with different grades of retinopathy. This finding is in contrast to the findings of Kreutzenberg
et al. [
27], who described that the retinopathy stage was associated with carotid atherosclerosis. Additionally, in the ARIC study, the severity of retinopathy was associated with cIMT. However, in contrast to the present study, patients with previous CV events were not excluded in those studies. Moreover, the absence of differences observed in cIMT measurements or in plaque presence in relation to retinopathy grades in the present study may be due to few patients in the group having more severe retinopathy. Thus, any conclusion drawn from these results must be made with caution.
In relation to retinopathy and the future risk of CV events, retinopathy has been described in population-based samples as a risk factor (independent from diabetes) for stroke [
47], heart failure [
48] and coronary heart disease [
49]. Patients from the ARIC study who had better CV health, according to the American Heart Association criteria, had a lower prevalence of retinopathy [
50]. Moreover, the presence of retinopathy and retinal microvascular signs has recently been described to be associated with an increased risk of cerebral microvascular disease [
51].
The strength of the present study is its specific design, which aimed to test the hypothesis that T2D patients with retinopathy have a higher atherosclerotic burden compared with T2D patients without retinopathy. Thus, a large number of patients with retinopathy were included. Only those patients with an eGFR >60 ml/min were included to avoid the confounding factor of renal insufficiency, which is associated with an increased risk of atherosclerosis.
The major limitation of the present study is that a direct relationship between retinopathy and macrovascular disease cannot be demonstrated due to the cross-sectional nature of the study. To assess the CV risk associated with a microvascular complication, such as DR in patients with T2D, longitudinal studies are warranted. To properly assess the CV risk, future studies should include T2D patients with DR (with and without carotid plaques) and T2D patients without DR and other microvascular complications.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
NA wrote and drafted the manuscript. AT and CJ performed the ophthalmological examination. All of the authors, except for DM, EF and EO, contributed to the data collection. EO participated in the study design, performed the statistical analysis and reviewed the manuscript. ER performed all ultrasound examinations. EF and DM conceived the study, participated in its design and coordination and reviewed the manuscript. All of the authors contributed to the discussion and approved the final manuscript.