Prognostic impact of carotid intima-media thickness and carotid plaques on the development of micro- and macrovascular complications in individuals with type 2 diabetes: the Rio de Janeiro type 2 diabetes cohort study

Participants and baseline procedures

This was a prospective study, the Rio-T2D Cohort Study, with 478 participants with type 2 diabetes enrolled between August 2004 and December 2008 and followed-up until December 2017 in the diabetes outpatient clinic of our tertiary-care University Hospital. All participants gave written informed consent, and the local Ethics Committee had previously approved the study protocol. The characteristics of this cohort, the baseline procedures and the diagnostic definitions have been detailed elsewhere [18, 24–28]. In brief, inclusion criteria were all adult type 2 diabetic individual (defined by the 2004 American Diabetes Association criteria [29]) up to 80 years old with either any microvascular (retinopathy, nephropathy or neuropathy) or macrovascular (coronary, cerebrovascular or peripheral artery disease) complication, or with at least two other modifiable cardiovascular risk factors. Exclusion criteria were morbid obesity (body mass index ≥ 40 kg/m²), advanced renal failure (serum creatinine > 180 μmol/l or estimated glomerular filtration rate < 30 ml/min/1.73 m²) or the presence of any serious concomitant disease limiting life expectancy. Specifically for this study, patients with previous carotid endarterectomy, previous neck irradiation, or with other possible etiologies of carotid disease, such as vasculitis or moyamoya disease, were also excluded. Diagnostic criteria for diabetic chronic complications were detailed previously [18, 24–28]. In brief, coronary heart disease was diagnosed by clinical, electrocardiographic criteria, or by positive ischemic stress tests; and cerebrovascular disease by history and physical examination. The diagnosis of nephropathy needed at least two albuminurias ≥ 30 mg/24 h or confirmed reduction of glomerular filtration rate (eGFR ≤ 60 ml/min/1.73 m², estimated by the CKD-EPI equation, or serum creatinine > 130 μmol/l). Peripheral neuropathy was determined by clinical examination (knee and ankle reflex activities, feet sensation with the Semmes–Weinstein monofilament, vibration with a 128-Hz tuning fork, pinprick and temperature sensations) and neuropathic symptoms were assessed by a standard validated questionnaire [25]. Clinic blood pressure (BP) was measured three times using a digital oscillometric BP monitor (HEM-907XL, Omron Healthcare, Kyoto, Japan) with a suitable sized cuff on two occasions 2 weeks apart at study entry. The first measure of each visit was discarded and BP considered was the mean between the last two readings of each visit. Arterial hypertension was diagnosed if mean systolic (SBP) ≥ 140 mmHg or diastolic BP (DBP) ≥ 90 mmHg or if anti-hypertensive drugs had been prescribed. Laboratory evaluation included fasting glycemia, glycated hemoglobin (HbA_1c), serum creatinine and lipids. Albuminuria was evaluated in two non-consecutive sterile 24-h urine collections. Laboratory examinations were repeated 2–4 times each year during follow-up, except albuminuria that was repeated once annually.

Carotid ultrasound imaging

A detailed description of carotid ultrasound measuring methods is available elsewhere [18]. In brief, a single experienced vascular radiologist, unaware of other participants’ data, performed all carotid ultrasound studies with a high resolution B-mode ultrasound (Sonoline G40, Siemens, Munich, Germany) and a 7.5 MHz linear array transducer. Carotid scanning protocol was that recommended by the Manheim Carotid Intima-Media Thickness Consensus [30]. Far-wall carotid IMT was visualized bilaterally at three sites: the common carotid artery (CCA-IMT, 20 to 60 mm from the flow divider), the carotid bulb (CB-IMT, 0 to 20 mm proximally from the flow divider) and the internal carotid artery (ICA-IMT, 0 to 20 mm distally from the flow divider). The images were digitally registered and carotid IMT measurements and plaque evaluation were performed off-line using an automated imaging processing software (Medical Imaging Applications, LLC Vascular Research Tools 5—Carotid Analyzer, Coralville, USA) managed by the sonographer. Measurements of three selected images at each site were performed and the mean IMT was calculated for each of the six locations. Left and right carotid IMT measurements were then averaged for the three sites. Carotid IMT measurement and plaque definitions were those recommended by the Manheim Carotid Intima-Media Thickness and by the American Society of Echocardiography Consensus [30, 31]. Plaque was defined as a focal structure that encroaches into the arterial lumen of at least 0.5 mm or 50% of the surrounding IMT value or demonstrates a thickness > 1.5 mm as measured from the media-adventitia interface to the intima–lumen interface [30, 31]. CIMT measurement did not include plaques. Grades of carotid stenosis were assessed by the velocity criteria and the real time B-mode images. A carotid artery plaque score quantifying method was also assessed [32]. Carotid artery segments (CCA, CB, ICA plus external carotid artery) were examined bilaterally. A grade was assigned to each examined segment: grade 0 for no observable plaque, grade 1 for 1 small plaque with diameter stenosis < 30%, grade 2 for 1 medium plaque with 30% to 49% stenosis or multiple small plaques, grade 3 for 1 large plaque with 50 to 99% stenosis or multiple plaques with at least 1 medium plaque, and grade 4 for 100% occlusion. The highest score observed in any carotid segment examined was considered the plaque score for each individual. The total number of arterial sites with plaques was also recorded. Intra-observer test–retest reliability of CIMT and plaque score in our laboratory had been reported previously [18]. The average intra-class correlation coefficient was 0.93 (95% CI 0.91–0.94, p < 0.001) and the ± 2 SD mean difference between the first and the second examinations varied from − 0.04 to 0.06 mm. Plaque score showed a good agreement with a hierarchical kappa value of 0.72 [18].

Follow-up and outcomes assessment

The participants were followed-up regularly at least 3–4 times a year until December 2017 under standardized treatment. The observation period for each individual was the number of months from the date of the first clinical examination to the date of the last clinical visit in 2017 or the date of the first endpoint, whichever came first. The primary endpoints were the occurrence of any cardiovascular event, specific microvascular outcomes and mortality. Total cardiovascular events (CVEs) were the following: fatal or non-fatal myocardial infarctions, sudden cardiac deaths, new-onset heart failure, death from progressive heart failure, any myocardial revascularization procedure, fatal or non-fatal strokes, any aortic or lower limb revascularization procedure, any amputation above the ankle, and deaths from aortic or peripheral arterial disease. Microvascular outcomes, previously defined, were the following: retinopathy development or worsening [26, 28]; a composite renal outcome [27, 28], defined as new microalbuminuria development or new renal failure development (defined as doubling of serum creatinine or end-stage renal disease [ESRD] needing dialysis or death from renal failure); and peripheral neuropathy development or worsening [25, 28]. Retinopathy and renal outcomes were evaluated by annual examinations [26–28], whereas peripheral neuropathy was evaluated on a second specific examination performed after a median of 6 years from the baseline examination [25, 28].

Statistical analyses

Continuous data were described as means (SD) or as medians (interquartile range [IQR]). Baseline characteristics of participants divided according to tertiles of CCA-IMT and to different plaque scores were compared by ANOVA, Kruskal–Wallis or χ² tests, when appropriate. Kaplan–Meier curves of cumulative endpoints incidence during follow-up, compared by log-rank tests, were used for assessing different incidences of outcomes among tertiles of CIMT and different plaque scores (0–1, 2, 3–4 points). For assessing the prognostic value of each carotid atherosclerosis parameters for each cardiovascular and microvascular outcome, except for peripheral neuropathy, a time-to-event Cox analysis was undertaken. First, analyses were only adjusted for age and sex, and then further adjusted for other potential confounders/risk factors (diabetes duration, body mass index [BMI], smoking, physical activity, diabetes treatment, arterial hypertension, number and classes of anti-hypertensive drugs in use, mean clinic SBP, presence of micro- and macrovascular complications at baseline, serum mean 1st-year HbA_1c, HDL- and LDL-cholesterol, and use of statins and aspirin). These results were presented as hazard ratios (HRs) with their 95% confidence intervals (CIs). For peripheral neuropathy analyses, a multiple logistic regression was used with the same statistical adjustments, except that height (instead of BMI) and the time interval between the baseline and second neuropathy evaluations were included as adjusting covariates. In all these analyses, CIMT was assessed as a continuous variable (with risks estimated for a 0.1 mm increment in CIMT) and also as two categorical variables (divided into tertiles with risks estimated for individuals in the highest tertile subgroup in relation to those in the lowest one, and dichotomized at clinically meaningful values: CCA-IMT > 1.0 mm, CB-IMT > 1.2 mm and ICA-IMT > 0.8 mm). Plaque score and number of vascular sites with plaques were both dichotomized at ≥ 3 points, with HRs estimated in relation to those individuals with values < 3. For assessing the improvement of discrimination performance after the addition of carotid parameters to a standard risk model, we used the C-statistic (analogous to the area under ROC curve applied to time-to-event analysis), compared by the method proposed by DeLong [33], and the integrated discrimination improvement (IDI) index [34, 35]. The IDI is equivalent to the difference in discrimination slopes between models with and without the new variable and its calculation is based on continuous differences in predicted risk in new and old models in individual cases and controls. Thus, the IDI is free of the dependence on empirical risk categories that is inherent to reclassification tables and can be used as an objective indicator of reclassification improvement. Both the absolute and the relative IDI were calculated. The relative IDI, reported as a percentage, facilitates the IDI interpretation, and is defined as the increase in discrimination slope divided by the slope of the standard model including only the traditional cardiovascular risk factors [34, 35]. In sensitivity and interaction analyses, interactions between carotid parameters and age (< 65 vs. ≥ 65 years old), sex, diabetes duration (< 10 vs. ≥ 10 years long), presence of micro- and macrovascular complications at baseline, and glycemic control (mean HbA_1c < 7.5% vs. ≥ 7.5%, < 58 mmol/mol vs. ≥ 58 mmol/mol) were tested for CVEs outcome and whenever there was evidence of interaction (p < 0.10 for interaction term), a further stratified analysis for that specific characteristic was performed. In all analyses a 2-tailed probability value < 0.05 was considered significant. Statistics were performed with SPSS version 19.0 (SPSS Inc, Chicago, Il., USA), and R version 3.4.1 (R Foundation for Statistical Computing, Vienna, Austria).

Baseline characteristics according to carotid atherosclerosis parameters

Endpoints occurrence during follow-up

After a median follow-up of 10.8 years (IQR: 8.7–12.0 years, maximum 13.3 years), 116 total CVEs occurred (corresponding to an incidence rate of 26 per 1000 patient-years of follow-up), and 115 individuals died (24 per 1000 patient-years), 56 from cardiovascular causes; 131 individuals newly-developed or worsened diabetic retinopathy, 156 achieved the renal composite endpoint (94 newly-developed microalbuminuria and 78 deteriorated renal function), and 83 individuals newly-developed or worsened peripheral neuropathy. Tables 1 and 2 show the incidence of each outcome respectively in participants divided by CCA-IMT and by plaque score. Individuals in the highest tertile subgroup of CCA-IMT and with the highest plaque score had an increased incidence of CVEs, all-cause mortality and cardiovascular mortality than those individuals in the lower CCA-IMT and plaque score subgroups. Kaplan–Meier curves of cumulative incidences over time (Fig. 1 for CVEs occurrence) confirmed these findings also for tertile subgroups of CB-IMT and ICA-IMT. Individuals in the highest plaque score subgroup also presented a higher incidence of renal endpoints, whereas the other microvascular outcomes incidences were not different in any subgroups of carotid atherosclerosis parameters.

Risks associated with carotid atherosclerosis parameters

Some evidences of interactions (p < 0.10) for CVEs occurrence were observed. There was interaction between plaque score and the presence of macrovascular disease at baseline (p = 0.09), where the higher plaque score was predictive of CVEs in those without prior cardiovascular diseases (HR: 2.57; 95% CI 1.31–5.07; p = 0.006), but not in those individuals with cardiovascular disease (HR: 1.45; 95% CI 0.78–2.71; p = 0.24). There was interaction between ICA-IMT and the presence of microvascular disease at baseline (p = 0.033), where ICA-IMT was a strong predictor of CVEs in individuals with microvascular disease at baseline (HR: 1.39; 95% CI 1.15–1.68; p = 0.001 and HR: 2.41; 95% CI 1.32–4.14; p = 0.004, respectively for continuous and dichotomized ICA-IMT in multivariate-adjusted analysis), but not in individuals without microvascular disease (HR: 0.95; 95% CI 0.75–1.21; p = 0.67 and HR: 1.08; 95% CI 0.54–2.17; p = 0.83). In further separated analyses, this interaction was mainly observed for diabetic retinopathy. Finally, evidence of interaction was also observed between ICA-IMT and glycemic control (p = 0.014), where ICA-IMT was a stronger predictor in individuals with lower HbA_1c levels (< 7.5%, 58 mmol/mol) (HR: 1.41; 95% CI 1.12–1.78; p = 0.003 and HR: 3.38; 95% CI 1.47–7.76; p = 0.004, respectively for continuous and dichotomized ICA-IMT) than in those with higher HbA1c (≥ 7.5%) (HR: 1.03; 95% CI 0.85–1.25; p = 0.73 and HR: 1.17; 95% CI 0.66–2.06; p = 0.60).

Improvement in risk stratification with carotid atherosclerosis parameters

Main findings

This prospective cohort study, with a median follow-up of 10.8 years, has some novel important findings. First, it demonstrated that CIMT measurement, particularly ICA-IMT, and presence of carotid plaques were significant predictors of renal outcomes in individuals with type 2 diabetes, but not of other microvascular complications development. Specifically, ICA-IMT predicted microalbuminuria development and renal function deterioration, whereas carotid plaques were mainly a predictor of renal function deterioration. Moreover, both increased ICA-IMT and plaque score improved risk discrimination for renal outcomes, with relative IDIs between 15 and 18%. Second, most CIMT parameters predicted adverse CVEs occurrence, but not all-cause or cardiovascular mortalities; and carotid plaques were significant predictors of cardiovascular morbidity only in individuals without cardiovascular diseases at baseline. Otherwise, the improvement in cardiovascular risk discrimination associated with carotid atherosclerosis was rather modest, with relative IDIs ranging from 1 to 8%. Overall, this study supports that ultrasonographic measurements of preclinical carotid atherosclerosis may be useful in the clinical management of patients with type 2 diabetes.

Carotid atherosclerosis and diabetic kidney disease

In longitudinal population-based studies, increased CIMT has been associated with progression of albuminuria [22], incident CKD [23] and incident ESRD [36]. However, longitudinal studies in patients with type 2 diabetes are scarce and with opposing findings [20, 21]. A previous study, with 162 individuals followed-up for 6 years, reported that CCA-IMT was associated with renal function deterioration [20]; whereas another study, with 1066 individuals followed-up for 6.7 years, reported no association between CCA-IMT and declining renal function [21]. None of them evaluated improvement in renal risk discrimination. We demonstrated here that CIMT measured at the internal carotid artery, but not at the more traditional common carotid segment, predicted adverse renal outcomes, microalbuminuria development and renal function deterioration, and improved renal risk discrimination by 15% to 18% in contrast to a standard risk factor prediction model. The number and stenosis severity of carotid plaques, quantified by a plaque score, also predicted renal function decline and improved renal risk discrimination by 18%. The reasons for the opposing findings between our study and the previous one [21] are not clear, but they may involve the methods and site of CIMT measurements, and differences in definition of renal function deterioration (in the previous study defined as eGFR < 60 ml/min/1.73 m² on two occasions with a > 25% decline from baseline, whereas in ours it was defined by doubling of serum creatinine to a value of at least 2.3 mg/dl or development of ESRD; hence, a more marked renal function decline). Otherwise, none of the previous studies [20, 21] evaluated the presence of carotid plaques as predictor of adverse renal outcomes. Also, a previous cross-sectional analysis demonstrated associations between urinary N-acetyl-β-d-glucosaminidase (NAG), a marker of renal tubular damage, and increased CIMT and plaques in type 2 diabetic patients [37], supporting pathophysiological links between diabetic kidney disease and carotid atherosclerosis. However, the present study adds important new data to current knowledge, because it included a relative large number of subjects, with longer follow-up (median of 10.8 years) and performed a comprehensive analysis including potential confounding factors, and further evaluated the improvement of the model with the addition of several ultrasonographic carotid parameters to renal outcomes prediction.

Carotid atherosclerosis and adverse cardiovascular outcomes

On the other hand, regarding CIMT and cardiovascular outcomes, our study confirmed previous ones [5–10, 38] by demonstrating that CIMT measured at different carotid segments, either analyzed as continuous variables or categorized at clinically meaningful cut-off values, mostly predicted future CVEs occurrence. However, they added only borderline modest contribution to cardiovascular risk discrimination improvement, being highest for carotid bulb measurements (from 6.6 to 8.4% improvement). This modest or negligible improvement in cardiovascular risk stratification after adding CIMT to traditional cardiovascular risk factors in individuals with type 2 diabetes has also been reported before [5, 6, 8, 10]. Regarding the prognostic value of carotid plaques for CVEs occurrence, dissimilar from a previous meta-analysis [11] where carotid plaques seemed a more powerful predictor than CIMT, in the present study the carotid plaque score was only borderline predictive of CVEs in the whole cohort, but it was a stronger predictor in the subgroup without pre-existent cardiovascular diseases at baseline, a fact not evidenced for CIMT. This observation supports the concept that, although both phenotypes, increased CIMT and presence of carotid plaques, may share some common mechanisms of initiation and progression, they represent distinctive grades and aspects of atherosclerosis [39–43]. Indeed, it has been demonstrated that they may have different prognostic importances in population-based studies [4, 44–46]. Otherwise, in our study, the carotid plaque score significantly improved cardiovascular risk discrimination, but with a relative improvement of 7.8%, comparable to that obtained by CIMT measurements.

Potential physiopathological mechanisms

One interesting observed interaction deserves mention. ICA-IMT was a stronger predictor of CVEs in individuals with pre-existent microvascular complications at baseline than in those without microvascular complications, particularly evident for diabetic retinopathy. This fact might, at least partially, underlie the reported associations between the presence of microvascular complications and higher cardiovascular risk [47–49], which seemed to increase linearly with the number of microvascular complications in a recent cohort study of individuals with type 2 diabetes [50]. We may also speculate that there might be a cross-talk between micro- and macrovascular disease in the pathophysiology of diabetic complications, where a pre-clinical marker of macrovascular atherosclerotic disease, such as increased CIMT or presence of asymptomatic plaques, predict future development of a microvascular complication (in this case, diabetic kidney disease); whereas, on the other hand, the presence of microvascular complications increase the risk of cardiovascular outcomes [47–50]. The pathophysiological links mediating this cross-talk between micro- and macrovascular complications possibly involve shared pathways such as insulin resistance, low-grade chronic inflammation, increased oxidative stress, and endothelial dysfunction [51]. Indeed, several adipose tissue-derived inflammatory factors were associated with the severity of carotid atherosclerosis in type 2 diabetic patients [52].