Natural history of vertebrobasilar dolichoectasia

Ectasia, elongation, and tortuosity of the basilar artery, vertebrobasilar dolichoectasia (VBD), is a condition characterized by a high degree of variability in the final outcome from one patient to another, with the full spectrum of the disease ranging from benign, even asymptomatic, to malignant. Clinical expression of VBD includes compression of cranial nerves or brainstem,^1–5 obstructive hydrocephalus,^3,6 ischemia in vertebrobasilar arterial territory,^7–13 and intracranial bleeding of various kinds.^7,11,14 Little is known about the natural history of this condition and even less is known about the progression of vascular abnormalities.^15–17

In this study a large cohort of patients with VBD was followed up to determine the long-term prognosis, identify predisposing factors for adverse events, and evaluate the frequency of VBD progression and its relation with clinical course.

METHODS

We prospectively collected information about 156 consecutive patients with VBD, hospitalized or seen as outpatients at the Department of Neurosciences between 1980 and 2000. Patients were followed until they died or until 2006. All patients were evaluated by CT scan (n = 24), MRI (n = 43), or both (n = 89). Fifty-three patients underwent additional angiographic examination and 46 MRA. In the remaining 57 patients, MRI was completely illustrative regarding VBD, so no additional vascular investigation was required. Other investigations included standard blood tests, Doppler ultrasonography of neck vessels, ECG, and transthoracic echocardiography. The basilar artery was considered elongated if at any point along its course it lay lateral to the margin of the clivus or dorsum sellae or bifurcated above the plane of suprasellar cistern. Ectasia was diagnosed if the minimum diameter of the basilar artery (BA) was greater than 4.5 mm.¹⁸ We only included patients with BA elongated and enlarged over its entire course without limitations as to age or presenting symptoms (figure 1). Elongation and uniform ectasia of other posterior circulation vessels corroborated the diagnosis of dolichoectasia in almost all patients. Patients with segmental, spindle shaped enlargements of the BA or fusiform enlargement superimposed on dolichoectasia, a condition recently described as a transitional type,¹⁶ were thus excluded, as were those with giant aneurysms (diameter greater than 25 mm).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Anteroposterior and lateral views of the vertebral angiogram showing slight (A, B) (diameter 5.9 mm, bifurcation in the suprasellar cistern) and severe (C, D) dolichoectasia (diameter 9.3 mm, bifurcation 3 cm above the dorsum sellae)

For the purpose of this study, the following concomitants were examined: history of hypertension (previous diagnosis of arterial hypertension: systolic blood pressure >160 mm Hg or diastolic >90 mm Hg or both or past or present use of antihypertensive agents), diabetes mellitus (previous diagnosis of diabetes or past or present use of antidiabetic agents), current smoking, alcohol abuse (>400 mL/week of pure ethanol), hyperlipidemia (cholesterol >250 mg/100 mL, triglycerides >180 mg/100 mL, or both), and history of coronary artery disease (CAD).

The diameter of the BA at midpons level, maximum diameter of the BA, height of bifurcation, and lateral displacement were evaluated as previously suggested.¹⁸ The height of the BA bifurcation was scored as 1 (within the suprasellar cistern), 2 (at level of third ventricle floor), and 3 (indenting and elevating the floor of the third ventricle); lateral displacement was scored as 1 (medial to lateral margin of clivus or dorsum sellae), 2 (lateral to lateral margin of clivus or dorsum sellae), and 3 (in cerebellopontine angle cistern) (figure 2). Scores of left and right displacement were averaged to obtain an index of laterality. We also measured the diameter of the vertebral arteries (VA), intracavernous internal carotid arteries (ICA), and of middle cerebral arteries (MCA) at the M1 segment. In absence of validated criteria for anterior circulation dolichoectasia, this condition was diagnosed on visual impression of tortuosity, elongation, and enlargement of these vessels and an ICA diameter >7 mm or an MCA diameter >4 mm. A VA diameter >4 mm was taken to indicate ectasia.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 MR examples for assignment of lateral position (left) and height (right) of basilar artery

(A) Medial to the lateral margin of clivus or dorsum sellae; (B) lateral to the lateral margin of clivus or dorsum sellae; (C) in the cerebellopontine angle cistern; (D) bifurcation in the suprasellar cistern; (E) bifurcation at the level of the floor of third ventricle; (F) elevating the floor of third ventricle.

Patients were classified as having atherosclerotic lesions (0 = no lesions; 1 = stenosis ≤30%; 2 = stenosis >30% and ≤70; 3 = stenosis >70%) of the anterior and posterior circulation on the basis of angiographic examination, MR angiography, and Doppler sonography.

Patients were followed up with periodic visits which included neurologic assessment, Doppler ultrasonography of neck vessels, and blood chemistry. Imaging studies were repeated if new symptoms or signs appeared or every 3 to 4 years in the absence of new events. Brain infarcts, intracranial bleeding, and other brain lesions were considered as such only when documented by imaging studies. Repeated imaging studies were compared to evaluate the presence of new brain lesions and the progression of VBD, i.e., increase in vertical elongation, lateral displacement, and diameter of the BA. Progression of VBD was defined as a change in diameter of the BA greater than 2 mm or as an increase in scores of lateral displacement or height of bifurcation of BA.

RESULTS

VBD progression.

Imaging follow-up was possible in 155 patients with an average duration of 10.1 years. Thirty-four patients underwent two, 53 patients underwent three, and 68 patients underwent four or more imaging studies. The average interval between imaging studies was 2.9 ± 1.1 years. During this period, VBD progressed in 67 patients (43%); this occurred within 5 years in 8 patients, within 10 years in 46 patients, and thereafter in 19 patients. Progression was more likely to be observed in younger patients (OR 0.953; 95% CI 0.923 to 0.984), when diameter of BA was larger (OR 1.257; 95% CI 1.015 to 1.555), score of height of bifurcation was high (OR 3.354; 95% CI 1.553 to 7.242), and when dolichoectasia also involved the anterior circulation (OR 2.530; 95% CI 1.227 to 5.219) (table 3). In the same period, serial imaging studies showed new brain ischemic lesion in 86 patients (55%), of which 59 were clinically evident and 27 asymptomatic. All silent ischemic lesions were of the lacunar type, whereas clinically evident ischemic lesions were of the lacunar type in 54 patients and of large territorial type in 5 patients. Progression of VBD was associated with all kinds of events, including death, and especially stroke (OR 8.50; 95% CI 4.09 to 17.67) (table 4).

View this table:

• View inline
• View popup
• Download powerpoint

Table 3 Baseline predictors of vertebrobasilar dolichoectasia progression in 155 patients

View this table:

• View inline
• View popup
• Download powerpoint

Table 4 Events during follow-up in relation to progression of vertebrobasilar dolichoectasia

Survival analysis found a cumulative proportion of survivors free of VBD related events of 54.1 (50.0 to 58.2) at 5 years, 39.5 (35.1 to 44.0) at 10 years, and 23.5 (18.2 to 28.9) at 15 years (figure 3). Survival curves were different between patients with and without VBD progression (log-rank test statistic 19.6; p > 0.0001). Patients who showed progression of VBD experienced at least one clinical event in the first 10 years after diagnosis in a proportion of 76% compared with 46% of patients without progression of VBD.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3 Event-free survival in the whole cohort (A) and in relation to the progression (filled circles) or no progression (empty circles) of vertebrobasilar dolichoectasia (B)

DISCUSSION

This study showed that, after the initial diagnosis, patients with VBD experienced clinical event possibly related to this condition, especially stroke, with a unexpectedly high frequency: 93 patients (60%) had at least one event in an average follow-up period of 11.2 years. Stroke was the most common clinical event (81% of all events) and was often recurrent. In our patients who presented with stroke, the cumulative 10-year risk of a first recurrent stroke was 56%. Although event rates obtained in different studies cannot be compared due to lack of standardization of age and gender, this figure exceeds that reported in a recent study, in which there was a 43% cumulative risk of recurrent stroke over 10 years in the general population,¹⁹ and is higher than the figure of 29% observed in Rochester, MN.²⁰ The high incidence of initial and recurrent ischemic stroke may reflect the evolution of the dolichoectatic process, as suggested also by the fact that 96 (98%) of the 98 brain infarcts we observed (41 at presentation and 57 during follow-up) were located in the vertebrobasilar arterial territory; however, it presumably also indicates incomplete efficacy of preventive treatment.

The high rate of recurrent ischemic stroke observed in our patients despite preventive treatment was in line with the observation of others,²¹ albeit in a small series of patients. This may be explained by the fact that there are various mechanisms by which VBD may promote transient or persistent brain ischemia, including distortion and stretching of the branches of the BA, reduction of anterograde flow in the dilated artery, and superimposed atheromatous changes.^12,22–24 Not all of these mechanisms are influenced by antiplatelet or anticoagulant treatment, which moreover may favor intracranial bleeding.¹⁴ As well, it has been suggested that dolichoectasia may also involve small perforating vessels.¹⁴ Indeed, a recent study reported an association between intracranial dolichoectasia and manifestations of small vessels disease in patients with stroke.²⁵ The fact that patients with VDB are more prone to lacunar infarcts than territorial infarcts could be a further factor limiting the efficacy of preventive treatment, because there is no specific secondary prevention trial demonstrating the efficacy of antiplatelet or anticoagulant therapy in small vessels disease. Finally, in our patients the factors associated with the occurrence of stroke coincided with factors describing the severity of dolichoectasia and its progression, none of which are modified by current preventive therapies.

During the long period of recruitment, both primary and secondary stroke prevention have progressed, especially with regard to risk factors management, as suggested also by the comparison between the two halves of the cohort (1980–1989 vs 1990–2000), which reflected a significant reduction in the prevalence of certain risk factors. However, this did not seem to influence the incidence of stroke as presenting symptom (47% vs 37%, p = 0.26) nor the cumulative proportion of recurrence at 5 years (55% vs 48%, p = 0.11) in patients presenting with stroke.

VBD may be progressive, as previously observed.^17,26 In our study progression of VBD was observed in 67 patients (43%) and was associated with a significantly increased morbidity and mortality. In a small sample of patients with VBD (n = 26), progression was observed in 23% of cases.¹⁷ This difference is probably related to the different duration of follow-up. Progression of VBD was more likely to be observed in younger patients, with a high degree of vertical elongation and greater diameter of BA and when dolichoectasia involved the anterior circulation: in other words, when the dolichoectatic process was already advanced at the time of diagnosis, especially in younger subjects. This is in line with the fact that progression of VBD tended to be more frequent in initially symptomatic patients (48%) than in nonsymptomatic patients (26%) (p = 0.042). However, symptoms at diagnosis did not correlate with VBD progression after controlling for BA measures, because symptomatic patients tended to have more severe VBD. The relation between initial diameter of aneurysm and its further enlargement has also been observed in patients with other kinds of nonsaccular aneurysms such as fusiform and transitional.^17,26 Progression of VBD influenced the likelihood of all types of clinical events, but the main effect was on the occurrence of ischemic stroke, which showed a sevenfold increase in risk and a tenfold increase in risk when silent brain infarcts were also considered.