Intracranial Spotty Calcium Predicts Recurrent Stroke in Patients with Symptomatic Intracranial Atherosclerotic Stenosis

This single-center prospective cohort study conformed to the ethical guidelines of the 1964 Declaration of Helsinki and its later amendments. All protocols were approved by the Institutional Medical Ethics Committee and informed consent was obtained from all participants. Symptomatic ICAS patients who were admitted to the Department of Neurology between January 2017 to June 2021 were consecutively recruited. The inclusion criteria were: (1) age 18–80 years old; (2) symptomatic ICAS referred to acute ischemic stroke in the anterior circulation identified by diffusion-weighted imaging; and (3) ≥ 50% stenosis on the relevant middle cerebral artery or intracranial internal carotid artery, as confirmed by magnetic resonance angiography, computed tomography angiography (CTA), or digital subtraction angiography. The exclusion criteria included: (1) patients with lacunar infarction; (2) coexistent ≥ 50% stenosis of the ipsilateral extracranial carotid artery; (3) non-atherosclerotic intracranial stenosis, such as dissection, vasculitis, moyamoya disease and reversible cerebral vasoconstriction syndrome; (4) evidence of cardioembolism (e.g., atrial fibrillation, mechanical prosthetic valve disease, sick sinus syndrome, dilated cardiomyopathy); (5) contraindications to CTA; and (6) patients received endovascular intervention during follow-up. All patients underwent CTA within 2 weeks of symptom onset to further evaluated the degree of stenosis and calcium characteristics. Patients with < 50% stenosis in the relevant intracranial artery on CTA imaging were also excluded from our study. Clinicians provided treatment for symptomatic ICAS patients according to the guidelines [21]. Patients with 50–69% stenosis and minor stroke were treated with 100 mg aspirin plus 75 mg clopidogrel for 21 days (followed by daily aspirin or clopidogrel alone) and control of stroke risk factors according to the CHANCE trial [22]. Patients with ≥70% stenosis were treated with dual antiplatelet therapy for 90 days according to the SAMMPRIS study [5].

All CTA examinations were performed using a dual-source 192-slice CT scanner (Somatom Force, Siemens Healthcare, Forchheim, Germany). Patients underwent non-enhanced CT imaging, followed by contrast-enhanced image scanning. The scanning parameters were set as follows: detector collimation 2 × 192 × 0.6 mm; gantry rotation time 0.25 s, pitch 3.2; tube voltage 70–90 kV; tube current by automated tube current modulation (CARE Dose4D, Siemens) using a reference tube current time of 330–450 mAs. Contrast media (Ultravist solution 370 mg I/mL; Bayer Healthcare, Berlin, Germany) was intravenously injected through an antecubital vein via a 20–22-gauge needle using a power injector. A total of 40–50 mL of contrast material was injected at a flow rate of 5 mL/s, followed by 60 mL of saline solution. Images were reconstructed using a slice thickness of 0.75 mm and an interval of 0.40 mm. Volume-rendered, maximum intensity projection, multiplanar reformatted, and curved planar reformatted images were generated to assess the carotid, cerebrovascular, and coronary arteries.

Calcium characteristics were evaluated by consensus review of two radiologists blinded to the clinical details. Atherosclerotic calcification was defined as a hyperdense region along the artery with CT attenuation ≥ 130 Hounsfield units (HU). The presence of calcification was measured in the coronary artery, aortic arch, extracranial carotid artery, and intracranial artery. Coronary artery included the left main, left anterior descending, left circumflex, and right coronary arteries. The aortic arch was assessed from the fused curved surface of the ascending and descending aorta to the origin of the left subclavian artery, the left common carotid artery, and the brachiocephalic trunk. The carotid artery was divided into C1–C7 segments according to the Bouthillier method [23]. The extracranial carotid artery was measured from the beginning of the common carotid artery to the C1 segment of the internal carotid artery. The intracranial artery was evaluated as the C2‑7 segment of the internal carotid artery, anterior cerebral artery, and middle cerebral artery. The calcium patterns were divided into SC and large calcific plaque according to the size of the calcific plaque. SC was defined as a calcific plaque with a maximum diameter of < 3 mm in any direction, and large calcific plaque was defined as a calcific plaque ≥ 3 mm in size [24]. Calcium density was measured on non-enhanced axial images with a slice thickness of 0.75 mm and a region of interest (ROI) of 0.5 mm². The highest CT value for each calcified plaque was recorded and defined as the calcium density by manually placing the ROI in all planes across the calcified plaque (Supplementary Fig. 1). The intracranial calcium was categorized into four groups according to quartiles of calcium density at baseline CTA imaging, with quartile 1 (Q1) for very low-density calcium (< 632 HU), quartile 2 (Q2) for low-density calcium (632–971 HU), quartile 3 (Q3) for high-density calcium (971–1316.5 HU), and quartile 4 (Q4) for very high-density calcium (≥ 1316.5 HU).

After CTA imaging at baseline, all patients were followed-up for at least 1 year. The clinical outcome was defined as recurrent ischemic stroke. All patients were followed up by telephone or routine medical outpatient clinic attendance and inquiring whether patients had experienced recurrent ischemic stroke in the past. If patients did not respond to the follow-up, we will find their medical records and try to contact their relatives for more information about the patient’s condition.

Continuous variables were presented as mean ± standard deviation or median (interquartile range, IQR). The Mann-Whitney U test and the t‑test were used to compare continuous variables between groups. Categorical variables were expressed as numbers (proportions) and compared using the χ²-test or Fisher’s exact test. Variables were included for multivariate analysis if they were p < 0.1 in the univariate analysis. Multivariable Cox proportional hazards regression was used to examine the association of calcium patterns and density with recurrent ischemic stroke. Cumulative event-free curves were constructed for recurrent ischemic stroke by the Kaplan-Meier method. All tests were 2‑sided, and p < 0.05 was considered statistically significant. All statistical analyses were performed with SPSS statistical software, version 25.0 (IBM, Armonk, NY, USA).

Of 230 patients with recent ischemic stroke in the anterior circulation and with ≥ 50% stenosis on the relevant intracranial artery, 196 were diagnosed with symptomatic ICAS. A total of 41 patients were excluded, including 16 patients who received preventive endovascular intervention during follow-up, 14 patients who were lost to follow-up, 7 patients with poor image quality, and 4 patients with contraindications to CTA examination. The flowchart of enrolled patients is shown in Fig. 1. A total of 155 patients (age 58.11 ± 11.64 years; 112 males) were included in the final analysis. The median follow-up was 22 months (IQR 19.6–24.3 months), During the follow-up, 29 patients (18.7%) experienced recurrent ischemic stroke. The clinical data are detailed in Table 1. Patients with recurrent ischemic stroke were older than those without recurrent ischemic stroke (62.93 ± 8.10 years vs. 57.00 ± 12.07 years, p = 0.027). No significant differences were found in other cardiovascular risk factors and lipid profiles between these two groups.

Of the 155 patients, 136 (87.7%) had at least 1 calcific plaque in the cardiovascular and cerebrovascular systems. The intracranial artery was the most frequently affected segment for calcification, followed by the coronary artery, aortic arch, and extracranial carotid artery. The presence of intracranial calcium was significantly higher in patients with recurrent stroke than those without (96.6% vs. 69.0%, p = 0.002, Table 2), whereas it was not statistically significant at the site of the extracranial carotid artery, aortic arch, and coronary artery. In addition, the incidence of intracranial calcium increased with age (p < 0.001), and no significant difference was found in the presence of intracranial calcium between patients with moderate and severe intracranial stenosis (65.9% vs. 77.2%, p = 0.155), or between the ipsilateral side and contralateral side to the stroke (70.3% vs. 63.2%, p = 0.185, Supplementary Fig. 2).

A total of 424 calcific plaques were found in intracranial arteries, of which 164 were SC and 260 were large calcific plaques. In the recurrence group, SC was found in 25 (86.2%) patients, including 8 patients with SC alone and 17 patients with both SC and large calcific plaques in intracranial arteries. In the non-recurrence group, SC was found in 51 (40.5%) patients, of whom 11 with SC alone and 40 with both SC and large calcific plaques (Table 3). Patients with intracranial SC were more likely to have recurrent ischemic stroke than those without (32.9% vs. 5.1%), with a hazard ratio of 7.03 (95% confidence interval, CI 2.44–20.23, p < 0.001). Furthermore, we compared the clinical characteristics according to the presence of intracranial SC. Patients with intracranial SC were older, more smokers, and had higher lipid levels (low-density lipoprotein cholesterol, triglycerides, and total cholesterol) than those without SC (Supplementary Table 1).

As shown in Table 4, the presence of very low-density calcium was higher in the recurrence group than in the non-recurrence group (72.4% vs. 37.3%), with the hazard ratio of 3.32 (95% CI 1.47–7.51, p = 0.001). On the per lesion level, large calcific plaque was denser than SC (1233.06 ± 297.39 HU vs. 687.93 ± 315.16 HU, p < 0.001, Supplementary Fig. 3).

As shown in Fig. 2, the Kaplan-Meier analysis with the log-rank test showed that participants without intracranial SC had a significantly higher recurrence-free rate than those with intracranial SC (p < 0.001). The variables with p < 0.1 in the univariable analysis, including age, prior stroke, the presence of intracranial calcium, the presence of intracranial very low-density calcium, and the presence of intracranial SC, were adjusted in the multivariable Cox regression analysis. There was no multicollinearity between variables in the multivariable analysis (the tolerance value > 0.1 and the variance inflation factor was < 10 for each covariate, Supplementary Table 2). After adjusting for confounding factors, the presence of intracranial SC remained significantly associated with recurrent ischemic stroke (adjusted hazard ratio 5.35, 95% CI 1.32–21.69, p = 0.019; Table 5). The increased stroke recurrence risk associated with intracranial SC remained significant in patients with severe intracranial stenosis (Supplementary Table 3), while neither the presence of intracranial calcium nor the very low-density calcium were significantly associated with recurrent ischemic stroke (p > 0.05).