Varying Definitions of Carotid Intima‐Media Thickness and Future Cardiovascular Disease: A Systematic Review and Meta‐Analysis

Authors: Yong Ling, Yiming Wan, Emma Barinas‐Mitchell, Akira Fujiyoshi, Hui Cui, Aikedan Maimaiti, Rong Xu, Jing Li, Chen Suo, Maryam Zaid

Abstract

Background

Carotid intima‐media thickness (cIMT) has been widely used as a predictor of future cardiovascular disease (CVD); however, various definitions of cIMT exist. This study provides a systematic review and meta‐analysis of the associations between different cIMT definitions and CVD.

Methods and results

A systematic review of the different cIMT definitions used in prospective cohort studies was performed. The relationships between cIMT of different definitions (common carotid artery IMT [CCA‐IMT], internal carotid artery IMT [ICA‐IMT], combined segments [combined‐IMT], mean CCA‐IMT, and maximum CCA‐IMT) with future stroke, myocardial infarction (MI), and CVD events were analyzed using random effects models. Among 2287 articles, 18 articles (14 studies) with >10 different cIMT definitions were identified and included in our meta‐analysis. After adjusting for age and sex, a 1‐SD increase in CCA‐IMT was associated with future stroke (hazard ratio [HR], 1.32 [95% CI, 1.27–1.38]), MI (HR, 1.27 [95% CI, 1.22–1.33]), and CVD events (HR, 1.28 [95% CI, 1.19–1.37]). A 1‐SD increase in ICA‐IMT was related to future stroke (HR, 1.25 [95% CI, 1.11–1.42]) and CVD events (HR, 1.25 [95% CI, 1.04–1.50]) but not MI (HR, 1.26 [95% CI, 0.98–1.61]). A 1‐SD increase in combined‐IMT was associated with future stroke (HR, 1.30 [95% CI, 1.08–1.57]) and CVD events (HR, 1.36 [95% CI, 1.23–1.49]). Maximum CCA‐IMT was more strongly related than mean CCA‐IMT with risk of MI, and both measures were similarly associated with stroke and CVD events.

Conclusions

Combined‐IMT is more strongly associated with CVD events compared with single‐segment cIMT definitions. Maximum CCA‐IMT shows a stronger association with MI than mean CCA‐IMT. Further research is warranted to validate our findings and to standardize the cIMT measurement protocol, as well as to explore underlying mechanisms.

Clinical PerspectiveWhat Is New? Combined intima‐media thickness (IMT), incorporating measurements from multiple segments within the carotid artery, exhibited superior performance compared with a single segment measure in the prediction of cardiovascular disease.Maximum common carotid artery IMT showed a stronger association with myocardial infarction compared with mean common carotid artery IMT. What Are the Clinical Implications? Combined IMT measurements may provide a more comprehensive reflection of the overall burden of atherosclerosis than those from single segments.Maximum common carotid artery IMT may be a more accurate representation of localized atherosclerotic plaques rather than the diffuse thickening of carotid arterial walls.The establishment of an optimized ultrasonographic protocol for the measurement and assessment of carotid IMT has the potential to enhance the predictive role of carotid IMT in cardiovascular disease risk estimation and can serve as a valuable tool in identifying individuals who may avoid developing future cardiovascular disease.

Cardiovascular disease (CVD) is one of the leading causes of morbidity and mortality on a global scale. ^1^ Atherosclerosis, the major underlying pathological process of CVD, is a chronic inflammatory state in which there is localized accumulation of lipids and inflammatory cells that make up atherosclerotic plaques. This process begins in early life and has a long latent and asymptomatic period before developing into a clinical condition, such as stroke, myocardial infarction (MI), and other CVD. ^2^

A well‐used method of assessing atherosclerosis at the subclinical stage is the measurement of the thickness of the intimal and medial layers of the carotid artery wall, termed carotid intima‐media thickness (cIMT). ^3^ The value of cIMT in research has been widely used as a surrogate measure of the burden of carotid atherosclerosis ^4^ and a predictor of CVD for primary prevention. ^5^ Meanwhile, the observed deceleration of cIMT progression resulting from therapeutic intervention implies a positive correlation with the reduction of CVD risk. ^6^ , ^7^ However, this has posed a major issue to the standardization of cIMT measurement protocol. ^8^ A careful analysis of cIMT research reveals considerable inconsistencies in cIMT measurement, including the carotid segments evaluated (common carotid artery [CCA‐IMT], internal carotid artery [ICA‐IMT], carotid bifurcation [bif‐IMT], or the combined segments [combined‐IMT]), ^9^ the measurement of the far or near walls of the segments, ^10^ the type of measurements made (mean or maximum of single measurements, mean of the mean, or mean of the maximum for multiple measurements), and whether or not plaques were included in the cIMT measurement. ^10^

Thus far, most studies had a preference for using CCA‐IMT ^11^ because measurements in this segment are more stable and reproducible in comparison to ICA‐IMT or bif‐IMT. ^12^ CCA‐IMT is more likely linked to systemic atherosclerosis and vascular remodeling in response to hemodynamic changes rather than ICA‐IMT, which is related to localized atherosclerosis. ^13^ Meanwhile, higher CCA‐IMT was found in individuals with prevalent stroke, whereas higher bif‐IMT was identified in those with prevalent coronary heart disease. ^14^ Systematic reviews and meta‐analyses have found that an increase in CCA‐IMT is more related to risk of future stroke than MI. ^15^ , ^16^ Moreover, there was evident heterogeneity in segment lengths measured and definitions did not overlap between studies. ^16^

This lack of standardization of cIMT was a major issue cited by the American College of Cardiology and American Heart Association for the routine use of cIMT for risk assessment in clinical practice. ^17^ The exact risks of future CVD in the general population associated with different definitions of cIMT, such as CCA‐IMT, ICA‐IMT, bif‐IMT, and combined‐IMT, are not entirely clear. To elucidate this, we performed a systematic review to identify the different definitions of cIMT used in prospective cohort studies and meta‐analysis to evaluate the association between different cIMT definitions and future CVD among the general population. We hypothesize that the inconsistent definitions used for cIMT across studies contribute to divergent associations between cIMT and future CVD outcomes.

Methods

The Meta‐analysis of Observational Studies in Epidemiology reporting guidelines were followed for conducting this systematic review. ^18^ The protocol for the review has been registered at the International Prospective Register of Systematic Reviews (www.crd.york.ac.uk/PROSPERO; CRD42022343324). The corresponding author can provide the data supporting the findings of this study upon reasonable request. As this systematic review and meta‐analysis used data from previously published studies, making ethics review or institutional review board approval not applicable. Informed consent was not required as we used available published data, with no access to any personal information of subjects.

Search Strategy and Study Selection

A thorough literature search in the medical literature databases MEDLINE, EMBASE, and Cochrane Library was conducted with the assistance of a medical librarian, using subject headings (MeSH and Emtree) and text words related to carotid intima‐media thickness and stroke or myocardial infarction and cohort from January 1950 to March 2022. Included studies had to be published and available in a complete English full‐text version and an additional review of the reference lists of reviews and meta‐analyses on cIMT was conducted to find more relevant articles. Authors were contacted if there were uncertainties concerning the studies. The complete literature search approach is given in detail in Data S1.

To be included in this analysis, studies had to meet the following participants had to be aged 18 years or older with no previous history of CVD during the baseline period; the exposure was defined as a 1‐SD or 0.1‐mm increase in cIMT measurement; comparisons were made between different values of cIMT measurements given that the exposure was a continuous variable; the outcomes of interest were stroke, MI, and CVD events (defined as combined end point stroke and MI); and the study design was limited to prospective cohort studies. Moreover, included studies needed to provide sufficient information and reliable data for evaluating hazard ratios (HRs) and 95% CIs.

Data Extraction and Quality Assessment

Two investigators (Y. L. and Y. W.) screened studies and extracted the data independently. Studies were screened first by title and abstract and then by full text. The following information of included studies was manually extracted (without use of data extraction software) by the investigators and recorded onto Microsoft Office Excel author's name, article title, study name, published year, country, population, sample size, mean age, % female, cIMT definition, end points, definition of end points, median follow‐up years, number of events, whether cIMT was modeled as continuous/tertiles/quartiles, adjusted variables, mean cIMT (mm), and HR with 95% CIs. The Kappa statistic was used to determine interrater agreement during the screening process. Any disputes were resolved via mutual agreement.

Newcastle‐Ottawa Scale (NOS) was used to assess the quality of included studies. NOS ranges from 0 to 9 stars including the quality of selection, comparability of the study and outcome determination. A NOS score >6 was considered a high‐quality study.

Definitions of cIMT

The segments of the carotid artery consist of CCA, ICA, bif, and external carotid artery, although the latter is not commonly used in measurements of cIMT. Studies identified in this meta‐analysis have used one segment or different combinations of the following 3 segments for the measurement of cIMT: CCA, ICA, and bif.

Several studies have used mean or maximum values of the CCA‐IMT segment measurements, whereas others have used ICA‐IMT, bif‐IMT, or a composite measure (we refer to as combined‐IMT) which is the mean of cIMT measurements from CCA, ICA, and bif ^19^ or 2 of the 3 segments (CCA and ICA ^20^ or CCA and bif ^21^ ).

Statistical Analysis

Forest plots were used to construct HRs and CIs for comparable studies. A random effects model recommended by Blettner et al. ^22^ with the method of moments approach from DerSimonian and Laird ^23^ was used to calculate pooled estimates if at least 2 studies reported sufficient data and a fixed effects model with the inverse variance method was used as a sensitivity analysis. The SEs of HRs were reevaluated by taking the logarithm of the CI points and dividing the resulting symmetrical CIs by 2 times the value of 1.96. ^16^ Heterogeneity was identified by a P value of <0.05 from the chi‐square test and an I ^2^ statistic by Higgins et al ^24^ greater than 50%. Publication bias with at least 10 studies included in the meta‐analysis was assessed using Begg's funnel plots and Egger's linear regression test. A nonparametric iterative trim‐and‐fill method recommended by Duval and Tweedy was used to calculate pooled estimates if Egger's test was significant. ^25^

To investigate the correlation between different definitions of cIMT, including CCA‐IMT, ICA‐IMT, bif‐IMT, and combined‐IMT, and the risk of future CVD, we performed a comparative analysis of age‐ and sex‐adjusted HRs and 95% CIs per 1‐SD increment and per 0.1‐mm increment. ^16^ Because CCA‐IMT is the most commonly used segment to assess the risk of asymptomatic CVD, some studies used mean values, whereas others used maximal values of CCA‐IMT. Therefore, we conducted separate meta‐analyses for mean CCA‐IMT and maximum CCA‐IMT to compare their associations with future CVD risk. Furthermore, we employed meta‐regression analysis with restricted maximum likelihood estimates, as recommended by Thompson and Sharp, to identify any possible factors contributing to significant heterogeneity in the meta‐analyses of at least 10 studies. ^26^

All analyses were performed using R language (version 4.1.2) with meta package. A 2‐tailed P value of <0.05 was considered significant.

Results

Description of Included Studies

A total of 18 articles that met our inclusion criteria were identified after the systematic review of 2287 published papers and the interrater agreement was high (Kappa=0.83). ^10^ , ^20^ , ^21^ , ^27^ , ^28^ , ^29^ , ^30^ , ^31^ , ^32^ , ^33^ , ^34^ , ^35^ , ^36^ , ^37^ , ^38^ , ^39^ , ^40^ , ^41^ The flow chart for the study is presented in Figure 1. An overview of the studies included in our meta‐analysis in terms of study population, sample size, definitions of different segments of cIMT, end points, follow‐up years, and model construction are depicted in Table 1. There were 17 publications reporting CCA‐IMT data, 2 publications reporting ICA‐IMT data, 1 publication reporting bif‐IMT data, and 3 publications reporting combined‐IMT data. Among the studies using CCA‐IMT, they were divided into 2 14 studies that used mean CCA‐IMT and 4 studies that used maximum CCA‐IMT. The Suita study was the only study that provided both mean CCA‐IMT and maximum CCA‐IMT in the same population. ^39^ The MDCS (Malmö Diet and Cancer Study), ^30^ , ^31^ SMART (Second Manifestations of Arterial Disease Study), ^32^ CVDFACTS (Cardiovascular Diseases Risk Factor Two‐Township Study), ^36^ PRC‐USA (People's Republic of China‐United States of America), ^21^ HNR (Heinz Nixdorf Recall), ^37^ , ^38^ and BLSA (Beijing Longitudinal Study of Aging) ^40^ studies made measurements of cIMT at sites free of any discrete plaques, whereas the Tromso Study ^35^ included the plaques in measuring cIMT. It is unclear what the other 7 studies did with plaques in their cIMT measurements. A schematic of the segments and definitions of cIMT used in each study can be seen in Figure 2, and 13 studies provided information on cIMT measurement protocol.

Study Quality Assessment

Table S1 summarizes the quality evaluation of the 18 articles based on NOS. All studies had sufficient exposure determination. Most studies included in our meta‐analysis were based on a general population without a history of stroke, MI, or CVD events. However, Lorenz et al enrolled participants with a history of stroke (1.9%) or MI (2.1%) at baseline, but subsequently excluded them from the analysis.

Relationship between cIMT and CVD

Meta‐analyses of the data adjusted for age and sex showed that the association of 1‐SD increment in CCA‐IMT with the risk of future stroke (HR, 1.32 [95% CI, 1.27–1.38]) was slightly stronger than the risk of MI (HR, 1.27 [95% CI, 1.22–1.33]). The association of 1‐SD increment in CCA‐IMT with the risk of CVD events (HR, 1.28 [95% CI, 1.19–1.37]) was weaker than that of stroke, but similar with that of MI (Figure 3). A 1SD increase in ICA‐IMT for predicting future stroke (HR, 1.25 [95% CI, 1.11–1.42]) and CVD events (HR, 1.25 [95% CI, 1.04–1.50]) were similar, but no significant association was observed for MI (HR 1.26 95% CI, [0.98–1.61]) (Figure 4). Due to the limited availability of data on bif‐IMT and CVD, specifically only 1 study, the association could not be evaluated. For combined‐IMT, an insufficient number of studies included data on MI, thus only stroke and CVD events were assessed. Per 1‐SD increment in combined‐IMT was associated with both future stroke (HR, 1.30 [95% CI, 1.08–1.57]) and CVD events (HR, 1.36 [95% CI, 1.23–1.49]) (Figure 5). A summary for the comparison of associations between per 1‐SD increment in different definitions of cIMT and risk of future stroke, MI, and CVD events is shown in Table 2. For 0.1‐mm increments in different cIMT definitions, significant associations were identified for CCA‐IMT and ICA‐IMT with stroke, MI, and CVD events. For combined‐IMT, CVD events maintained an association, whereas stroke (HR, 1.06 [95% CI, 0.98–1.14]) did not reach statistical significance (Figures S1–S3). Substantial and considerable heterogeneity (I ^2^ > 50% and P value <0.05) was identified in some associations (Figures 3C, 4B, 4C, and 5A; Figures S1, S2B, S2C, S3A, and S3B).

Mean and Maximum CCA‐IMT

Studies evaluating CCA‐IMT were grouped and compared by measurement methods that used mean CCA‐IMT or maximum CCA‐IMT measurements. Age‐ and sex‐adjusted HRs per 1‐SD increment in mean CCA‐IMT were 1.32 (95% CI, 1.27–1.36) for stroke, 1.25 (95% CI, 1.19–1.32) for MI, and 1.27 (95% CI, 1.17–1.37) for CVD events. The HRs for maximum CCA‐IMT were 1.30 (95% CI, 1.14–1.48) for stroke, 1.35 (95% CI, 1.23–1.47) for MI, and 1.25 (95% CI, 1.09–1.44) for CVD events (Table 3, Figures 6 and 7). Similar results were found for per 0.1 mm increment in mean and maximum CCA‐IMT, albeit with a lower effect size. However, the relationship between maximum CCA‐IMT and CVD events did not demonstrate statistically significant association (Figures S4 and S5). Substantial and considerable heterogeneity (I ^2^>50% and P value <0.05) was identified in some associations (Figures 6C, 7A, and 7C, Figures S4A–S4C, S5A, and S5C).

Publication Bias and Sensitivity Analyses

No evidence of publication bias was found for per 1‐SD increment of CCA‐IMT with CVD events (P value of Egger's test=0.68) or mean CCA with CVD events (P value of Egger's test=0.44), or per 0.1‐mm increment of CCA‐IMT with CVD events (P value of Egger's test=0.96) or mean CCA with CVD events (P value of Egger's test=0.90) (Figure S6). Similar estimates were found in the fixed effects models with narrower CIs (Figures S3–S7, Tables S2 and S3, Figures S1–S5). The diverse cIMT definitions and CVD all showed significant associations, whereas some of these did not attain statistical significance in the random effects models.

Meta‐Regression Analysis

Meta‐regression analysis investigating the moderating effects of sample characteristics (mean age, proportion of women, baseline cIMT value, follow‐up duration, and ethnicity) and measurement properties (far versus far and near wall IMT, right versus bilateral sides, and position and length of the examined cIMT segment) revealed that the duration of follow‐up was the strongest and the only significant source of heterogeneity in the pooled estimates of age‐ and sex‐adjusted HRs for stroke per 0.1‐mm increment in mean CCA‐IMT (P=0.023) (Figure S7).

Consensus Protocol for Measurement of cIMT

Among 14 studies in 18 articles, 1 study (BLSA) adopted the Mannheim consensus and measured mean cIMT in plaque‐free areas on the far wall of the distal CCA‐IMT, defined as at least 5 mm before the dilation of the bulb, of the left and right carotid arteries as the value of cIMT measurement. Three studies (ARIC [Atherosclerosis Risk in Communities], CVDFACTS, and PRC‐USA) adopted the ARIC protocol, which measured the mean cIMT in plaque‐free portions of the far wall of the 3 segments (CCA, ICA, and bif) in both right and left arteries. Two studies (SMART and Suita) used mean cIMT in plaque‐free areas of the far wall or both far and near walls of the CCA, defined as the point 10 mm proximal to the beginning of the dilatation of the carotid bulb, of left and right carotid arteries as the value of cIMT measurement (Table 4). Nevertheless, the remaining 8 studies used their own protocol for measurement of cIMT.

Discussion

We identified >10 different measurement protocols for cIMT used in prospective cohort studies of general populations. Although published data for some cIMT definitions were limited, our meta‐analysis revealed that a composite measure of cIMT using 2 or more segments of the carotid tree (ie, combined‐IMT) had a stronger association with overall CVD events than the often‐used CCA‐IMT. Moreover, our comparison of mean and maximum CCA‐IMT suggests that maximum CCA‐IMT may have a stronger association with future MI.

Segment‐Specific cIMT and CVD

Similar to our findings, a previous meta‐analysis found that CCA‐IMT was more related to risk of stroke than MI. ^15^ The British Regional Heart Study demonstrated that higher CCA‐IMT was related to prevalent stroke but not to ischemic heart disease. ^14^ In our study, CCA‐IMT was also associated with higher risk of MI and CVD events, and compared with ICA‐IMT, CCA‐IMT had a stronger relationship with stroke. This result is inconclusive as data from only 2 publications were available on ICA‐IMT. Moreover, there were a limited number of publications to assess the association strength of bif‐IMT for future CVD. Later stages of atherosclerosis (plaque, stenosis, and occlusion) are common at ICA and bif locations and occur only occasionally in CCA. ^14^ , ^42^ Historically, because measuring CCA‐IMT with ultrasound was believed to be more stable and reproducible than measuring ICA or bif, ^12^ studies had a preference for using CCA‐IMT. ^43^ However, the development of automated edge‐tracking software has improved the reproducibility of cIMT measurements by eliminating the need for manual measurements. ^44^

Combined‐IMT and CVD

From the studies that met the screening criteria, there were several definitions of combined‐IMT that used mean of the mean IMT values or mean of the maximum IMT values for (1) CCA and ICA, (2) CCA and bif, or (3) CCA, bif, and ICA. Noticeably, combined‐IMT had a stronger relationship with CVD events compared with CCA‐IMT. For the relationship of combined‐IMT with stroke and MI, there were discernably fewer publications, although stroke was found to have a positive association. Combined‐IMT also had a stronger relationship with stroke and CVD events in comparison to ICA‐IMT, although the number of studies were limited for both cIMT measures. The IMPROVE (Carotid Intima Media Thickness and IMT‐Progression as Predictors of Vascular Events in a High‐Risk European Population) study demonstrated that combined‐IMT had a stronger association for combined cardiovascular or cerebrovascular events than individual carotid artery segments, especially mean of maximal values in carotid arteries. ^45^ Because plaques are more likely to develop at ICA and bif locations, ^11^ measurements of combined‐IMT may reflect the general atherosclerosis burden better than those from single segments. ^45^

Mean CCA‐IMT, Maximum CCA‐IMT, and CVD

As we have demonstrated, cIMT has varying definitions in terms of segments and their lengths (Figure 2, Data S2). However, another critical source of variation is the mathematical description of the thickness. Mean and maximum of several measurements have both been used to describe cIMT, especially CCA‐IMT. The IMPROVE study found that mean CCA‐IMT had a slightly stronger relationship with the risk of CVD events than maximum CCA‐IMT. ^45^ In contrast, we found that mean and maximum CCA‐IMT had similar strengths of association for stroke and CVD events, whereas the HR for MI was noticeably higher for maximum CCA‐IMT. Mean CCA‐IMT may represent a condition of generalized wall thickening, correlating to both hypertension‐related stress and atherosclerotic plaques. ^12^ In comparison, maximum CCA‐IMT may better represent focal atherosclerotic plaques ^12^ rather than diffuse thickening of carotid arterial walls, explaining the stronger association with MI. ^8^ There is a lack of careful consideration of the meaning of cIMT and what each type of measure represents. Mean measurements are more common than maximum measurements and, interestingly, a substantial number of studies that use mean measurements specifically measure cIMT in plaque‐free areas of the carotid arteries. ^45^

Consensus Protocols

Although cIMT has been used extensively in cardiovascular research over the years, there is a lack of standard protocol for how this technology should be used as a research tool. Some study groups, however, have attempted to create cIMT consensus protocols. The Mannheim consensus, established in 2004 and amended in 2006 and 2011, proposed that cIMT be measured preferably on the far wall of the CCA at least 5 mm below the bulb within an area free of plaque with a clearly identifiable double‐line pattern. ^46^ The American Society of Echocardiography consensus statement published in 2008 recommended that cIMT measurements should be restricted to the far wall of the CCA and supplemented by a thorough scan of the extracranial carotid arteries for the presence of carotid plaque to increase sensitivity for identifying subclinical vascular disease. ^47^ In addition, The ARIC protocol recommends using mean IMT on the far wall of the 3 segments (CCA, ICA, and bif) in both the right and left arteries. In our meta‐analysis, 1 study adopted the Mannheim consensus, 3 studies adopted ARIC protocol, but no other established consensus was followed by the remaining studies. This is certainly a major source of heterogeneity. In fact, the 2013 Cardiovascular Risk Clinical Practice Guidelines highlighted that having a standardized measurement was a major challenge and that routine measurement of cIMT was not advised in clinical practice for risk assessment for a first atherosclerotic CVD event. ^17^ Standardization of the definition and measurement of cIMT is not only critical for its potential application as a screening tool but also for its use as a surrogate marker of CVD risk in epidemiologic research.

Study Strengths and Limitations

Our study has several strengths. To the best of our knowledge, this is the first systematic review and meta‐analysis to assess the strength of association of different definitions of cIMT with future CVD. There are currently few studies that use carotid segments other than CCA. In particular, studies describing the association between ICA and bif and future CVD are scarce. Importantly, we have identified that most prospective cohort studies in general populations use incongruent definitions of cIMT. The findings from this study may serve as motivation for the future standardization of cIMT measurements, especially with the evident inconsistencies in cIMT segment lengths and locations we have highlighted. Moreover, all studies included in the analysis were evaluated using NOS and demonstrated a relatively high level of quality.

Nevertheless, our study also had several limitations. The first limitation of our study is the potential publication bias. Due to the low numbers of included studies in most meta‐analyses, it is difficult to estimate the risk, which may alter the results. Second, heterogeneity between studies may have affected our results. We performed meta‐regression analysis where follow‐up duration was revealed to be a significant source of heterogeneity; however, meta‐analyses of <10 studies with substantial heterogeneity were not applicable for a meta‐regression to address. Moreover, additional heterogeneity of the source data, including minor differences in inclusion criteria and CVD end point definition, cannot be entirely ruled out. Even within the same segment and type of cIMT measurement, most studies had varying measurement methods (eg, angles, walls, segment lengths), which further contributes to heterogeneity. Third, other segments and measures, such as ICA‐IMT, maximum CCA‐IMT, and at times combined‐IMT were limited to 2 studies per category, resulting in limited statistical power to reasonably estimate HRs. Fourth, the exclusion of articles published in languages other than English may have led to the omission of certain studies. Finally, our results are primarily derived from 1‐SD increments in different cIMT definitions and we did not center our conclusions on results from 0.1 mm increments as such a dimensioned quantity is not appropriate for comparisons. This is because cIMT segments (ie, CCA‐IMT, ICA‐IMT, and bif‐IMT) have diverse base values and ranges.

Conclusions

Our findings indicate that combined‐IMT is more strongly associated with CVD events as compared with cIMT definitions that use only 1 segment, such as CCA or ICA. Additionally, maximum CCA‐IMT appeared to exhibit a stronger association with MI than mean CCA‐IMT. However, it is essential to acknowledge that the discrepancies in cIMT measurements across studies could affect the observed associations. Moreover, given the limited available research, it is premature to draw a final conclusion on the strength of associations of different definitions of cIMT with future CVD. It is crucial for further research in larger and more diverse populations to enhance the generalizability of the results, to standardize cIMT measurement protocols, as well as to investigate the underlying mechanisms linking different cIMT definitions with CVD.

Sources of Funding

This work was supported by The Local High‐Level Discipline Construction Project of Shanghai. The funder played no role in the design and conduct of the study nor any stage of the writing.

Disclosures

None.

Supporting information