Abstract
Background: Cervical artery dissection is often attributed to an underlying arteriopathy related to a generalized extracellular matrix defect.
Objective: The authors compared the hemodynamic and morphologic properties of the carotid artery, as assessed noninvasively by ultrasonography, in patients with spontaneous internal carotid artery dissection (ICAD) and control subjects.
Method: Twenty-six patients who experienced ICAD more than 6 months before evaluation were compared with 26 controls matched for age, sex, and height. Cases and controls had ultrasound measurement of common carotid artery diameter and diameter change during the cardiac cycle, bulbar and suprabulbar internal carotid artery diameters, and common carotid artery intima–media thickness. The unaffected carotid artery in cases was compared with the carotid artery of the same side in controls.
Results: Common carotid artery relative diameter change was significantly higher in cases than controls, whereas other measurements were not significantly different between the groups. In multivariate analyses, the highest tertile of common carotid artery relative diameter change was associated with the risk of ICAD (OR, 10.0; 95% CI, 1.8 to 54.2; p = 0.002)
Conclusion: An underlying arteriopathy, presumably related to an extracellular matrix defect, may be present in patients with spontaneous ICAD.
Spontaneous internal carotid artery dissection (ICAD) , one of the main causes of ischemic stroke in young adults,1 is now easily identified by carotid ultrasonography and MRI. The pathophysiology of ICAD is unknown, but it is postulated that patients with ICAD have a predisposing underlying arteriopathy related to an extracellular matrix defect.2 This notion is supported by indirect evidence, such as associations with heritable connective tissue disorders,3 intracranial aneurysms,4 aortic root dilatation,5 reduced type III collagen production by dermal fibroblasts,6 ultrastructural abnormalities of dermal connective tissue components,7 and clinical examination abnormalities suggesting a connective disease.8
Hemodynamic and morphologic properties of large elastic arteries are determined mainly by the extracellular matrix components of the arterial wall9,10⇓ and can be evaluated noninvasively by ultrasonography.11,12⇓ We compared the hemodynamic and morphologic ultrasonic properties of common (CCA) and internal (ICA) carotid arteries in patients with spontaneous ICAD and controls.
Patients and methods.
Patient selection.
We reviewed the medical records of 69 consecutive patients (mean age [SD] = 47.1 [8.8]; 32 women) seen in our department between 1989 and 1996 with extracranial ICAD. Inclusion criteria for the current study were a spontaneous unilateral ICAD that had occurred more than 6 months before evaluation and the absence of any characteristics that might modify the morphologic and functional evaluation of carotid arteries. Patients with bilateral ICAD (n = 10), previous contralateral ICAD (n = 1), residual ICA stenosis (i.e., stenosis above 50%) or occlusion (n = 14) on follow-up imaging study performed 2 to 6 months postdissection, head or neck trauma in previous days or weeks (n = 3), or recent ICAD (n = 3) were not contacted for participation in the study. Among the 38 remaining patients, five were lost to follow-up, one had died, and eight were being followed at another institution and declined to participate. The other 24 agreed to enter the study. Two patients who were part of the group regularly followed up in our institution and met the inclusion criteria, but had experienced ICAD before 1989, were also included.
Thus, the current study included 26 patients (17 men, 9 women; mean age [SD] = 49.7 [8.0], range 35 to 69) with a spontaneous unilateral ICAD that had occurred more than 6 months before evaluation, as confirmed by MRI and MR angiography (n = 19) or conventional angiography (n = 7). Controls were recruited among healthy volunteers of the hospital staff with no history of cervical arterial dissection. Each case was matched with one control for age (±5 years), sex, and height (±10 cm). For each subject (patient and control), the medical history focused on connective tissue disorders, current medications, cigarette smoking, and oral contraceptive use. Blood pressure and cardiac pulse measurements were recorded.
Ultrasound examination.
The study was carried out using a B-mode ultrasound imaging system (Acuson 128 XP, Mountain View, CA) with a 7.5-MHz transducer (axial resolution: 0.4 mm) and a bidirectional Doppler device (Angiodyne DMS; Perols, France) equipped with a spectrum analyzer and a 4-MHz continuous wave Doppler probe. All ultrasonic investigations were performed by two experienced observers (P.J.T., V.A.) blinded to clinical data but not to status (patient or control). Patients were in supine position in a darkened room, and the examination was begun after at least 10 minutes of rest.
The CCA was insonated from the neck behind the sternocleidomastoid muscle. Ultrasonic B-mode images of both CCA (2 cm before the bifurcation), ICA bulbs, and the suprabulbar segment of ICA were obtained in a longitudinal section, and echoes from the vessel wall were optimized. The transducer was positioned perpendicular to the vessel inner wall to obtain the largest diameter of each arterial segment, and frozen images were recorded on a videotape (S-VHS). In addition, a dynamic sample (at least 5 seconds) was also recorded on each CCA. Postprocessing of the ultrasonic signals was performed off-line using computer-assisted image analysis (M’Ath; Metris, Argenteuil, France). The reader was not aware of the status (patient or control). A line was drawn with the computer mouse on fixed images perpendicular to the vessel axis. Based on density changes from the lumen to the vessel wall structures, the largest end-diastolic cross-sectional intraluminal diameter of the carotid bulbs and the suprabulbar segment of ICA was automatically measured. CCA interadventitial diameter and the intima–media thickness of the CCA proximal wall were similarly measured.13 All measurements were repeated three times, and the mean retained. The continuous variation of interadventitial diameter throughout two to four cardiac cycles was measured on recorded dynamic samples by drawing a 10 to 25 mm line along the vessel longitudinal axis. The maximum interadventitial systolic diameter (Ds) and the minimum interadventitial diastolic diameter (Dd) were automatically calculated and selected from the values obtained on all measurements done on 25 images per second. The pulsatile diameter change (Ds − Dd) was displayed on the screen. Each measurement was repeated twice, and the mean value recorded. Common carotid artery relative diameter change, adapted from arterial strain [(Ds − Dd)/Dd], was defined as [(Ds − Dd)/Dd] × 100%.14
Statistical analysis.
Cases and controls were first compared with regard to general characteristics and ultrasound variables. Data were obtained from the unaffected carotid artery (i.e., the side contralateral to the dissection) in cases and from the carotid of the same side in controls. Variables significantly associated with ICAD were then entered together into the same logistic model, and a stepwise backward procedure was applied. Crude and adjusted odds ratios (OR) were used as a measure of the association between carotid artery ultrasonographic variables and risk of ICAD. Analyses were carried out using SAS (release 6.12) (Cary, NC).
Results.
At the time of dissection diagnosis, the mean age (SD) of the 26 patients was 46.9 (8.1). The dissections (right ICA in 11 patients and left ICA in 15) were all considered nontraumatic, even though trivial activity involving head movements (such as sports activities) or recent infection had occurred in six cases in the days preceding the first symptoms. Signs and symptoms consisted in various combinations of pain (n = 24), Horner’s syndrome (n = 15), cerebral or ocular ischemia (n = 12), cranial nerve palsy (n = 2), and tinnitus (n = 1). No patients had a history of connective tissue disorder. The mean time elapsed between ICAD and ultrasound study was 34.2 (37.0) months.
The baseline characteristics of cases and controls are presented in table 1. Both groups were comparable for age, height, and body mass index. There were eight current smokers and two current oral contraceptive users in the control group versus four and one in the dissection group (differences not significant). None had a history of cardiopulmonary disease or diabetes. Mean systolic and diastolic blood pressure were higher in cases, but the difference was significant only for diastolic blood pressure.
Baseline characteristics of cases and controls
Morphologic data are presented in table 2. In cases, CCA diameter was lower and bulbar and suprabulbar ICA diameters higher than in controls, although these differences did not reach significance. The only significant difference between cases and controls concerned CCA relative diameter change, which was higher in the former (11.6% versus 10.3%; p = 0.018). To analyze this association further, the relative proportion of cases and controls was estimated by tertiles of CCA relative diameter change. There was a regular decrease of the proportion of controls and a regular increase of the proportion of cases by increasing tertiles of CCA relative diameter change (figure). The association between CCA relative diameter change and risk of ICAD was then assessed in a logistic model in which the lowest tertile of CCA relative diameter change was used as the reference. This showed an increase of risk of ICAD by increasing tertiles of CCA relative diameter change; i.e., the highest OR were observed for the highest tertile of CCA relative diameter change (table 3). When other anatomic characteristics of the carotid artery—mainly common, bulbar, and suprabulbar carotid artery diameter—were included in the model, the risk of ICAD for the highest tertile of CCA relative diameter change was even greater (adjusted OR, 14.9; 95% CI, 2.2 to 101.7).
Ultrasound characteristics of cases and controls*
Figure. Relative frequency of cases (black bars) and controls (gray bars) by increasing tertiles of common carotid artery (CCA) relative diameter change. Tertiles of CCA relative diameter change were defined as less than 9.5% (tertile 1), 9.5 to 11.8% (tertile 2), and more than 11.8% (tertile 3).
Risk of carotid artery dissection expressed in tertiles of CCA relative diameter change
Discussion.
This study shows that CCA relative diameter change was increased in patients with spontaneous extracranial ICAD in comparison with matched controls. In multivariate analysis, the risk of ICAD associated with the highest tertile of CCA relative diameter change, as compared to the lowest tertile, was 10.0 (1.8 to 54.2). Other morphologic variables, such as intima–media thickness, CCA diameter, and suprabulbar and bulbar ICA diameters, were not significantly different between cases and controls.
The current study attempted to provide a reliable evaluation of the morphologic and functional characteristics of carotid arteries in patients with ICAD. As the healing process within the dissected vessel may modify arterial dynamic and morphologic properties, strict limitations on entry criteria were used and analyzes were restricted to data recorded on the nondissected side. Patients were examined at least 6 months after occurrence of ICAD, and those with significant residual significant stenosis and occlusion (which could modify cervical hemodynamic conditions) or bilateral ICA dissection (i.e., without an unaffected carotid artery) were excluded.
Certain methodologic limitations may have affected our findings. The observers performing the ultrasound examination could have been aware of the status (case or control) of the subjects examined, although they were unaware that analyses would be performed on the data obtained from the nondissected side and that the hypotheses tested concerned arterial distensibility. It is therefore unlikely that these unblinding conditions had any influence on the recording of ultrasonographic measurements. Controls were recruited among healthy volunteers of the hospital staff and therefore rather selected. However, their general characteristics did not differ significantly from those of patients, and it is rather unlikely that this selection bias could have had any influence on the difference in arterial distensibility observed between cases and controls. Finally, the sample was rather small, which could limit the power of comparison between cases and controls. With a larger sample, some of the differences observed—for example, in suprabulbar carotid artery diameter—could become significant. Nonetheless, this would not modify the main result for CCA relative diameter change, which was already significant.
Other factors with influence on arterial distensibility (such as vascular risk factors or heart and valve dysfunctions) might have been under-recognized in our study. However, none of our patients had a history of cardiovascular disease and vascular risk factors mainly act on distensibility by the way of arteriosclerosis. Intima–media thickness, a reliable marker of arteriosclerosis,15,16⇓ showed similar values in both groups, indicating that the influence of these factors might be limited. Furthermore, the similar value for intima–media thickness in both groups is consistent with the notion that cervical dissection results from a nonatheromatous arteriopathy, as suggested by rare autopsy findings.17,18⇓
CCA relative diameter change is an index of arterial distensibility, which is a one-dimensional assessment reflecting the elastic properties of the arterial wall.11,14⇓ These elastic properties are related to a passive component from elastic and collagen fibers and to an active component from smooth muscle fibers.19 As the CCA has a high proportion of elastic fibers and only a small quantity of smooth musculature (elastic type artery), vessel wall movements are mainly related to the relative amounts of elastin and collagen and their anatomic relationship.10,12⇓ Thus, altered CCA distensibility may be considered as an indicator of disturbed vessel wall integrity and hence a possible predictor of vessel wall fragility. Distensibility was only assessed in the CCA because this artery is more easily and accurately accessible to ultrasound than the ICA. Although dissection rarely occurs in the CCA, it is likely that the properties of this artery are quite similar to those of the ICA, as the entire extracranial carotid artery tree is of the elastic type and contains the same blood vessel components.
The increased distensibility found in patients with ICAD reinforces the hypothesis of an underlying arteriopathy related to an extracellular matrix defect. A number of anomalies indicative of an extracellular matrix defect have been described in patients with ICAD, such as arterial redundancies,20 intracranial aneurysms,4 aortic root dilatation,5 reduced type III collagen production by dermal fibroblasts,6 ultrastructural abnormalities of dermal connective tissue components,7 and symptoms of heritable connective tissue disorders.8 The exact nature of these dissection-associated anomalies remains unknown, although a qualitative or quantitative deficiency in the fibrillar component, endogenous proteolytic hyperactivity, and protease inhibitor hypoactivity have been proposed.21-24⇓⇓⇓
Acknowledgments
Supported by a grant from Collège des Enseignants de Neurologie/Laboratoire Lafon (B.G.) and ADNLA.
Acknowledgment
The authors are grateful to all patients and volunteers who participated in the study.
Footnotes
-
Presented at the 52nd annual meeting of the American Academy of Neurology; San Diego; April–May, 2000.
- Received March 2, 2000.
- Accepted May 25, 2000.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.