Barzin et al. [1] published in the recent issue a study titled “Comparative analysis of adolescent hypertension definitions for predicting early-adulthood carotid artery intima-media thickness: Tehran lipid and glucose study.” The main objective of the study was to investigate the prevalence of hypertension (HTN) among adolescents according to 3 different accepted definitions for childhood HTN (for details the reader should refer to the original study) [1]. Barzin et al. [1] compared the performance of each of the 3 definitions in predicting a high carotid intima-media thickness (CIMT), used as surrogate marker for preclinical atherosclerosis. The study showed a difference in CIMT between 4 blood pressure (BP) groups only when 4th report definition was applied, revealing a significant difference between the high-normal BP and normal BP (0.58±0.09 mm vs. 0.55±0.09 mm, P<0.05) [1]. The prospective cohort study included 921 individuals aged 10–17 years [1]. Participants were categorized into normal BP, high-normal BP, HTN stage 1, and HTN stage 2 groups based on the childhood HTN definitions of the 4th report, European Society of Hypertension, and American Academy of Pediatrics Clinical Practice Guidelines [1]. From the methodological description it appears that the authors measured the CIMT for most patients at one predetermined section, the far wall of the common carotid artery (CCA) and for a subset of subjects also at the internal carotid artery (ICA) and the carotid bifurcation. That is not plausible as it introduces a major methodological heterogeneity into the measurements rendering the results between the included subjects not comparable: Intermediate stages occur between increased CIMT and atherosclerotic plaque (for ultrasound not distinguishable) and that occur commonly at the bifurcation and the origin of the ICA and less at the CCA, where generally the measurements occur [2]. Furthermore, it is unclear from the methodological description from which side and site the CIMT value was used: initially the authors stated that an average CIMT value obtained from 3 locations of both CCAs was used, only to finish that the CIMT from the left CCA was utilized.

If only the left side was used, importantly to mind and what the authors need to consider for their results, that left side CIMT are reported higher than right side CIMT [2]. A further important source of measurement heterogeneity was introduced by the authors by not synchronizing with the cardiac cycle, namely the end-diastolic phase [2-4]. Using CIMT as surrogate marker has advantages, as it is a safe, noninvasive and cost-effective examination. Yet, CIMT is a delicate biomarker and requires high-precision measurements, as submillimetric differences are sufficient to categorize subjects into different CIMT categories upon which important clinical as well pathophysiological deductions are formulated. There are 2 major recommendations as to CIMT protocols: one is to perform measurement at the far wall of the CCA (foremost for the higher spatial resolution) [2] and the second is to perform a composite CIMT measurement including different segments of the carotid artery (CA), e.g., CCA, bulb and ICA [5-8]. The latter will arguably capture the asymmetric nature of atherosclerosis [9,10] in a more accurate manner respect a single CIMT measurement location. However, whatever the pros and cons of these 2 recommended measurement protocols [2,5] are, it is of paramount importance to apply for the same cohort the same measurement protocol to guarantee the uniformity of measurements of the included subjects. The cardiac synchronization is of critical importance as vessel diameter and consequently CIMT are subject to obvious variations during the cardiac cycle with reported mean CIMT differences between the 2 cardiac phases of 0.037 to 0.041 mm [3,4]. In conclusion: When CIMT as surrogate marker is used an utmost detailed measurement protocol is required, and all procedural steps need reporting. The CIMT results and related conclusions of this study should be considered within the above-mentioned methodological limitations [1].