Blood vessel tortuosity is a common vessel anomaly affecting a range of vessels, from large arteries to small arterioles, in a number of organ systems. Severe tortuosity may cause a variety of serious symptoms (1). Twisted or tortuous arteries have often been reported in cerebral and internal carotid arteries (ICAs) (1), and the prevalence of tortuosity was significantly higher in the ICA system than in the vertebrobasilar system (2). However, a very different prevalence of ICA tortuosity has been reported (3, 4). The disagreement in prevalence may be due to difference in study cohort and diagnostic technique and lack of diagnostic criteria. Several possible measures of tortuosity have been suggested, but none of them have obtained universal acceptance. Tortuosity index (TI) is commonly used as an index of tortuosity because it was a relatively straightforward approach in the absence of diagnostic criteria for anterior circulation vessel tortuosity (2). TI value can be easily measured without special software. In addition, in clinical applications, these values can be rapidly obtained to evaluate the degree of vascular tortuosity.

Cerebral small vessel disease (SVD) refers to all pathological processes that affect the small vessels of the brain, including small arteries and arterioles, capillaries, and small veins (5). There are six closely correlated features in SVD that can be observed on brain magnetic resonance imaging (MRI), including recent small subcortical infarcts, white matter hyperintensities (WMHs), lacunes, cerebral microbleeds (CMBs), enlarged perivascular spaces (EPVSs), and atrophy (6). An association between artery tortuosity and MR neuroimaging markers of SVD has been reported, especially in the posterior circulation (7–9). However, a few studies involved the whole MRI spectrum of SVD in association with ICA tortuosity, while majority of the SVD lesions occur in the basal ganglia and periventricular region (10).

In our previous study, we found that intracranial artery calcification correlated with the neuroimaging findings of SVD, which suggested a certain relation between large vessel disease and SVD. Based on these findings, we hypothesized that large artery tortuosity was associated with SVD. The present study aimed to investigate the correlation between ICA tortuosity and the neuroimaging markers of SVD using craniocervical artery computed tomography angiography (CTA) and to determine if ICA tortuosity is a risk factor for SVD and whether it can act as a marker of disease severity.

The interobserver agreement was found to be acceptable. The values for the presence of CMBs, EPVSs, WMHs, and lacunes were 0.936, 0.928, 0.912, and 0.909, respectively (all P < 0.001).

The baseline characteristics of the study population (n = 1,264) are shown in Table 1. Patient age and prevalence of hypertension, diabetes mellitus, current smoking, and previous stroke significantly differed between patients with and without SVD MRI spectrum (P < 0.001, respectively; Table 1). White matter hyperintensities were found in 47.4% of the subjects, lacunes in 47.3%, CMBs in 32.8%, and EPVSs in 27.0%. Univariate analysis showed that the ICA TI value was positively associated with the presence of each SVD marker (all P < 0.001; Figures 2, 3). The correlation coefficients (r_s) were 0.57, 0.42, 0.30, and 0.26 for EPVSs, WMHs, CMBs, and lacunes, respectively (Table 2). The crude ORs of the ICA TI were 1.59 (95% CI 1.53–1.66, P < 0.001) for EPVS grade 1, 2.24 (95% CI 2.12–2.37, P < 0.001) for EPVS grades 2–4, 1.18 (95% CI 1.13–1.23, P < 0.001) for WMH grades 1–2, 1.62 (95% CI 1.53–1.71, P < 0.001) for WMH grade 3, 1.23 (95% CI 1.18–1.28, P < 0.001) for lacunes, and 1.31 (95% CI 1.25–1.37, P < 0.001) for CMB. After adjustment for age, sex, diabetes mellitus, hypertension, hyperlipidemia, current smoking, previous stroke, prior coronary artery disease, atrial fibrillation, and three other SVD MRI markers (when one of the four neuroimaging markers was analyzed), the adjusted ORs of the ICA TI were 1.52 (95% CI 1.44–1.60, P < 0.001) for EPVS grade 1, 2.05 (95% CI 1.93–2.18, p < 0.001) for EPVS grades 2–4, 1.01 (95% CI 0.97–1.05, P = 0.614) for WMH grades 1–2, 1.09 (95% CI 1.03–1.15, P = 0.004) for WMH grade 3, 0.97 (95% CI 0.94–1.01, P = 0.161) for lacunes, and 0.98 (95% CI 0.93–1.02, P = 0.233) for CMB (Table 3 and Figure 4).

In the current study, the TI of ICA was found to be higher in patients with the neuroimaging of SVD. After adjustment for the confounding factors, ICA tortuosity was associated with MRI-defined markers of SVD, including EVPSs and high-grade WMHs; furthermore, ICA tortuosity was positively correlated with EPVS severity.

Our findings suggested that ICA tortuosity was significantly correlated with EPVSs. In the previous studies, arterial tortuosity was reported to correlate with the neuroimaging of SVD (17–20), while only a few studies have addressed EPVSs. Pico et al. (7) found that patients with ischemic stroke and intracranial arterial dolichoectasia (IADE) have a higher prevalence of EPVSs. However, the diagnosis for IADE was made by consensus based on the visual impression and without diagnosis criteria. Moreover, the study only involved three imaging markers of SVD (WMHs, EPVSs, and lacunes). In the present study, we used a TI value to evaluate the severity of ICA tortuosity because it was a relatively straightforward approach in the absence of diagnostic criteria for anterior circulation vessel tortuosity. Furthermore, we investigated the correlation between ICA tortuosity and four imaging markers of SVD, since several MRI features are considered to often co-occur and might be confounding factors to each other (15). We assumed that there were several possible explanations for our findings. Blood vessel tortuosity was a systemic problem and could also affect small vessels (1). Brain small vessel tortuosity could lead to stiffer arteries (21) and transmit greater pulsatility to the brain parenchyma (22), which might lead to enlargement of perivascular spaces. Additionally, the tortuous cerebral arterioles slightly extend and wind on themselves, while pushing back into the perivascular space. Simultaneously or subsequently, there is a loss of brain parenchyma around the tortuous vessel, resulting in an EPVS to permit further tortuosity of the blood vessel (18). This might be a long and dynamic process, so we also found that ICA tortuosity was positively correlated with EPVS severity. These findings are interesting because the pathophysiology of EPVS is unclear, and these results need to be confirmed in further studies.

Additionally, we found that ICA tortuosity was only correlated with the high-grade WMHs (grade 3) after adjustment for potential confounding factors. Our results were not completely consistent with the results of previous studies. These inconsistencies may be due to differences in study cohort and methodology. Some studies established an association between IADE and WMHs (7–9). However, one was a brain-autopsy study concerning fatal stroke patients (9), and others concerned posterior circulation arterial tortuosity (7, 8). Some studies also reported a higher prevalence of lacunar infarctions in the patients with IADE and stroke (23). Park et al. (24) found that vertebrobasilar dolichoectasia in patients with TIA or stroke was associated with CMBs. However, they investigated posterior circulation arterial tortuosity in stroke patients, and the sample size was rather small. In the present study, the individuals were suspected of having cerebral vessel diseases, not only the patients with stroke. Individuals who had large area infarction, hemorrhagic stroke, and ipsilateral artery stenosis >50% were also excluded. Furthermore, this study only concerned ICA tortuosity. Therefore, our findings might be somewhat different from those reported in other studies. Further studies including anterior and posterior circulation arterial tortuosity are needed to confirm the results.

Previous studies mainly analyzed the relationship between IADE and SVD; however, there are many tortuosity cases that do not involve dilatation. In addition, Smoker's criteria provided a relatively consistent means to diagnose posterior circulation IADE (25); however, there were no diagnostic criteria for anterior circulation IADE (2). Several possible measures of tortuosity have been suggested, but none of them have obtained universal acceptance. Furthermore, anterior and posterior circulation arterial tortuosity does not coexist in many cases, which may due to different structures surrounding the vessels. Therefore, in the present study, we only measured the ICA tortuosity and used TI to evaluate the severity of tortuosity because of its ease of computation and conceptual simplicity.

This study has several limitations. First, the severity of atherosclerosis is different among individuals; however, almost 20% of IADE patients do not have traditional cardiovascular risk factors, suggesting that IADE can occur without atherosclerosis (25). Second, our controls did not represent “healthy” controls because completely healthy populations cannot undergo CTA; we selected the less severe side of the ICA for analysis when the degree of tortuosity of both ICAs was not consistent. However, both conditions may lead to an underestimation of the association between the SVD severity and TI value. Third, it is difficult to conclude the grading standard for ICA tortuosity in this study because we only analyzed the patients with SVD MR features, and the sample size of the patients with EPVSs was relatively small. Last, the current study had a retrospective design, which was a substantial limitation.

In conclusion, the TI of ICA was higher in patients with neuroimaging of SVD. Internal carotid arteries tortuosity was associated with SVD MRI-defined markers, including EVPSs and high-grade WMHs. Internal carotid arteries tortuosity was positively correlated with EPVS severity. Arterial tortuosity might be a risk factor for SVD with implications for understanding the pathophysiology of SVD. This finding may have potential clinical significance for identifying individuals with suspected SVD and evaluating disease severity.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Shanghai Jiao Tong University Affiliated Sixth People's Hospital institutional review board. The patients/participants provided their written informed consent to participate in this study.

Y-HL and Y-CC contributed conception and design of the study. Y-CC, R-HQ, and X-FS organized the database. X-EW and JL performed the statistical analysis. Y-CC wrote the first draft of the manuscript. Y-HL, Y-CC, X-EW, and JL wrote sections of the manuscript. All authors contributed to manuscript revision, read and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.