Carotid intima‐media thickness as a target‐organ damage and treatment‐target: Need for a major revision?

Arterial hypertension emerged as a predominant health care aspect over the last few decades and the management of patients with high blood pressure (BP) has been presented as a crucial challenge in everyday clinical practice.1 Hypertensive individuals are at increased risk of cardiovascular morbidity and mortality. All-cause mortality and elevated BP levels detrimentally affect several target-organs.2, 3 Specifically, hypertension leads to functional and structural abnormalities in the myocardium, brain, kidneys, vasculature, and retina. The assessment of several indexes for hypertension-related target-organ damage, such as left ventricular hypertrophy and left ventricular mass, urinary albumin excretion and estimated glomerular filtration rate, pulse wave velocity, pulse pressure, ambulatory arterial stiffness, carotid intima-media thickness (CIMT) and coronary artery calcification, ankle-brachial index, as well as sexual dysfunction revealed that a subclinical target-organ damage could exist years prior to the manifestation of a major cardiovascular event (ie, myocardial infarction, stroke, cardiovascular death).3, 4 These findings suggest that the evaluation of target-organ damage could offer a significant prognostic value among hypertensive patients. Nowadays, it is well established that the presence of subclinical target-organ damage independently magnifies cardiovascular risk. However, even if the evaluation of target-organs of hypertensive patients in total is recommended,3 it remains uncertain (with few exceptions) whether BP-lowering therapy could improve the above mentioned deteriorations, and more importantly, whether these ameliorations result in clinically meaningful benefits (reduction in major cardiovascular events) or a well-controlled hypertension status.5 Thirty-years ago, Pignoli et al.6 first described in-vivo measurement of arteries intima-media thickness via ultrasound. Probably due to the anatomical position of carotid arteries, CIMT is widely used in clinical practice and cardiovascular studies as an indicator of atherosclerotic damage. For many years, changes over CIMT had been used as a surrogate cardiovascular endpoint, however, this adoption seems quite arbitrary.7 Indeed, it is well established that increased CIMT is associated with elevated BP levels, resulting in more prevalent cardiovascular events in both hypertensive and normotensive individuals.3, 8-10 Nevertheless, for CIMT to be considered as a real cardiovascular risk factor and not a marker or an epiphenomenon, a change in CIMT due to a therapeutic intervention has to be related with similar cardiovascular benefit.11 In the 2-year Cholesterol Lowering Atherosclerosis Study (CLAS), the treatment with colestipol-niacin led to attenuation of CIMT increase rate.12 The study cohort (146 males who previously underwent coronary artery bypass grafting) was assessed after an average of 8.8 years of follow-up following the integration of CLAS. Interestingly, it was found that a lower rate of CIMT increase resulted in a reduction in coronary events. In contrast, the European Lacidipine Study on Atherosclerosis (ELSA) reported that baseline CIMT level is related with cardiovascular events, but no predictive value of CIMT (on any major cardiovascular endpoint) was observed when the analysis included baseline and on-treatment CIMT values.13 The effects of treatment-induced CIMT changes on cardiovascular risk have been further evaluated in a few meta-analyses, providing conflicting outcomes. In a meta-regression analysis of 15 598 patients, it was documented that regression in CIMT progress between active-treatment (antihypertensive, antidiabetic, or lipid-lowering) and placebo groups was associated with a greater risk of myocardial infarction in the placebo-arm.14 Specifically, every 0.01 mm/y regression in CIMT features a lower risk of myocardial infarction (odds ratio [OR]: 0.82; 95% CI: 0.69 to 0.96). Of note, this finding was not confirmed among randomized controlled trials assessing statin treatment; an observation that could suggest a potential strong role of antihypertensive treatment in such benefits. Costanzo et al.15 conducted a meta-regression analysis to evaluate the effect of treatment-induced change in CIMT on cardiovascular events, by pooling data from 41 studies (n = 18 307). It was found that CIMT regression was not assessed with significant reductions in myocardial infarction (τ2: 0.91; P = .37), cerebrovascular events (τ2: −0.32; P = .75), and all-cause mortality (τ2: −0.41; P = .69). Likewise, Lorenz et al. reported the outcomes of an individualized data meta-analysis, which failed to provide a correlation between CIMT progression and cardiovascular hard endpoints (hazard ratio [HR]: 0.98; 95% CI: 0.95 to 1.01).16 However, with a closer look on the design of the abovementioned meta-analyses, several inherent limitations seem to restrict the generalization of the findings.7, 17 In brief, significant flaws include the pooling of trials with wide heterogeneity in relevance with the risk stratification and baseline CIMT values of participants, administered medication, ultrasound protocols, adjudication of events, and CIMT definition and calculation. Additionally, meta-regression analyses are underpowered to detect relationships and along with ecological fallacy could further limit the outcomes in the case of meta-analyses by Goldberger et al. and Costanzo et al.14, 15 Recognition of these flaws lead van den Oord et al.18 to conduct a meta-analysis with a more appropriate design. A total of 15 studies that measured CIMT via B-mode ultrasound were pooled in this meta-analysis. The outcomes suggest that a 1 SD increase in CIMT significantly elevates the risk of myocardial infarction by 26% and stroke by 31%. Moreover, elevation in CIMT of 0.1 mm resulted in augmentation of myocardial infarction and stroke risk (HR: 1.15; 95% CI: 1.12 to 1.18 and HR: 1.17; 95% CI: 1.15 to 1.21, respectively). More recently, in a meta-analysis of more than 3000 adults younger than 45-years and free of symptomatic cardiovascular disease, it was noted that an increase in CIMT by 1 SD is associated with the first myocardial infarction or stroke event (HR: 1.40; 95% CI: 1.11 to 1.76).19 Accumulating data suggest that antihypertensive treatment could blunt CIMT progression. Puato et al.20 studied 20 young hypertensive subjects by measuring CIMT with a B-mode ultrasound, bilaterally. It was concluded that a well-controlled BP results in minimization of CIMT progression (Δ: 0.084 mm; P = .004). ELSA was the first randomized, controlled trial that evaluated the effect of BP-lowering therapy on CIMT progression and till now remains the only study that assessed whether this outcome improves cardiovascular events.13 Head-to-head comparison of lacidipine vs atenolol unveiled that the calcium-channel blocker (CCB) was associated with a greater deceleration of CIMT progress than the beta-blocker, but as discussed previously, these outcomes were not associated with clinically significant cardiovascular benefits. A sub-study of the Hypertension Control Based on Home Systolic Pressure (HOSP) pilot study evaluated the impact of strict (< 130 mm Hg) vs usual (130-139 mm Hg) home systolic BP control with either amlodipine or losartan among the 80 hypertensive participants.21 No significant alterations in CIMT progression were found in strict- vs usual-group. Amlodipine resulted in an initially greater attenuation of CIMT progression compared with losartan, but this disparity did not exist after the fifth year of follow-up. Meta-analytic data indicate that long-term antihypertensive therapy significantly ameliorates CIMT progression.22, 23 Regarding the efficacy of different antihypertensive agents on deceleration of CIMT progression, Cuspidi et al.22 reported that CCB achieved a greater reduction than beta-blockers, diuretics, and angiotensin converting-enzyme (ACE) inhibitors. Tropeano et al.23 noted that ACE inhibitor treatment was associated with a reduction of 0.08 mm in CIMT vs placebo (95% CI: −0.14 to −0.02). Further analysis revealed a significant greater benefit of CCB vs ACE inhibitors as well as of angiotensin-receptor blockers over beta-blockers on CIMT. In addition, diuretics were less efficacious than CCB on improving CIMT. In the current issue of the journal, Kim et al. publish a cross-sectional study of 1809 Korean participants to assess differences in CIMT level between hypertensive and normotensive subjects as well to investigate potential disparities in CIMT value between different antihypertensive drug classes.24 Approximately 60% of the study sample suffered from hypertension. These hypertensive patients were more likely to present higher body mass index, be older, be of male gender, have diabetes mellitus, and have dyslipidemia. They were also more likely to be treated with statins compared to normotensive individuals. It was also shown that hypertensive participants had a greater CIMT, carotid intima thickness, and carotid media thickness compared with normotensive participants (0.72 ± 0.17 vs 0.64 ± 0.15; 0.31 ± 0.07 vs 0.30 ± 0.06; 0.41 ± 0.13 vs 0.35 ± 0.12 mm, respectively; P < .001 for all comparisons). Additionally, beta-blockers in the hypertensive group were associated with increased CIMT (0.03 mm; P = .018), carotid intima thickness, and carotid median thickness compared with hypertensive patients not treated with beta-blockers. No correlation between other antihypertensive drug classes and CIMT was noted. Last, after multivariable analysis, only carotid intima thickness was substantially associated with beta-blockers (odds ratio: 2.49; P = .016). The large sample of participants and the well-structured methodology of measurements of CIMT are significant strengths of the study by Kim et al.24 However, the trial carries several limitations. First, several baseline characteristics, comorbidities, and concomitantly administered medications could be significant confounding factors and may substantially affect study findings. Second, the cross-sectional design of the study carries the inherent inability to identify causal relationships since it does not include pre- and post-treatment values. Collectively, the study design casts doubts about the clinical meaning of study outcomes, especially when considering the lack of data about the duration of antihypertensive treatment and the type of administered beta-blockers (atenolol vs vasodilatory agents). In summary, CIMT is a significant indicator of cardiovascular risk. BP-lowering therapy substantially attenuates CIMT progression and CCB, and in a lower-grade renin-angiotensin system, blockers seem to exert greater benefit. However, it remains inconclusive whether this amelioration of CIMT progression has a simultaneous cardio-protective effect. Future long-term, large-sample, randomized controlled trials are needed to investigate whether BP lowering therapy per se, or in combination with other medications, including antidiabetic and hypolipidemic drugs, could result in significant cardiovascular benefits. The authors report no specific funding in relation to this research and no conflicts of interest to disclose.