Neurology -- Correspondence for Pico et al., 63 (11) 2016-2021

Correspondence published:

Reply to Ubogu et al: Pierre Amarenco, Fernando Pico, Ariel Cohen (11 May 2005)

Intracranial arterial dolichoectasia is associated with enlarged descending thoracic aorta: Eroboghene E. Ubogu, Osama O. Zaidat (11 May 2005)

Reply to Ubogu et al 11 May 2005

Pierre Amarenco,
Department of Neurology and Stroke Centre
Bichat University Hospital and Medical School, 46 rue Henri Huchard, Paris 75018, France,
Fernando Pico, Ariel Cohen
Send Correspondence to journal:
Re: Reply to Ubogu et al

pierre{at}amarenco.net Pierre Amarenco, et al.

We thanks Dr Ubogu and Zaidat for the interest. [1] They raised concerns regarding the method we used to define dolichoectasia in our cohort. We disagree with most of their comments and the criteria to which they refer.
Our cohort included 510 consecutive patients with brain infarction on MRI. Since there are no validated criteria, we chose to use the consensus method (i.e., visual impression) that there was at least one dolichoectatic artery (IADE+ patients). The important point is not the IADE diagnosis method used, since there are no validated criteria, but to characterize IADE+ patients and IADE- patients. We also used a 16-diopter lens to systematically measure the diameter of seven intracranial arteries on MRI (both carotid siphons, both middle cerebral arteries, both vertebral arteries and basilar artery). We used the box-plot method and found that 94% of IADE+ patients actually were in the highest quartile of diameter distribution of one of the seven arteries. Except for the basilar artery (BA), we could not find a threshold value to define dolichoectasia. These results have been reported in a previous paper. [2]
We also read with interest the paper by Drs. Ubogu and Zaidat. [3] They used a 4-diopter lens to measure three intracranial arteries on MRA, and then used an arbitrary 4.5 mm threshold value to define dolichoectasia. Both MRA and the above mentioned threshold value have never been validated for this diagnosis. This value was based on Smoker�s paper in which 126 patients with normal CT-scans had BA measurement. [4] Mean BA diameter was 3.17 mm and the 2 SD upper limit was 4.5 mm. This value is obviously dependent on the sample and is not associated with a clear threshold effect on the risk of stroke or death. This is the reason why this threshold value should not be used as such in other cohorts. Limitations of threshold values for continuous variables have been demonstrated with blood pressure or serum cholesterol levels, for example. In addition, MRI luminal measurements are smaller than CT external diameter measurements of BA.
Therefore, it is inadequate to say that 15% of our IADE+ patients actually had dolichoectasia, following the arbitrary threshold value used by Ubogu and Zaidat, because this assumption is only based on BA diameters, although dolichoectasia may involve another intracranial artery without involving the BA. This is another reason why we used the consensus method: there are no validated criteria (threshold value) to define dolichoectasia of end carotid artery, middle cerebral artery or vertebral artery. The consensus method was therefore the least bad method to define dolichoectasia (IADE+ patients) in our cohort, provided we did characterize these IADE+ patients by measuring seven intracranial arteries, and further noted that 94% of IADE+ patients were in the fourth quartile of artery diameter distribution of at least one of the seven arteries.
In Ubogu and Zaidat�s paper, the mean diameter and distribution (SD, minimum and maximum values) of BA (and vertebral artery) in cases and controls were not mentioned. Box-plots were not displayed in the paper. Furthermore, it is not shown whether external diameter of BA on CT was equivalent on MRA. They also used combined diagnosis criteria (diameter or deviation). We do not know how these criteria were distributed in their sample or if the majority of patients were diagnosed vertebrobasilar dolichoectasia because of diameter or because of length or deviation, or because both criteria were present.
Ubogu and Zaidat suggested to measure the diameter of 11 intracranial arteries rather than 7 that we measured, but they chose to measure only three arteries in their paper.
Contrast angiography is no longer, if ever, the gold standard to measure aortic diameter. It has possible side effects and does not allow to visualize the vessel wall. TEE allows to measuring thoracic aorta diameters with excellent accuracy. [5, 6] We did not define a priori a threshold value for aortic enlargement. We rather used aortic diameter as continuous variable and then divided into quartiles, which is the standard, strongest method.
We compared aortic diameters to other measured intracranial artery diameters and found similar results. However, we used comparison with only BA diameters because they had the best intra-reader reproducibility and because BA was the most often ectatic artery.
After adjustment on confounding factors, p values for the association between BA, ascending and thoracic descending aorta diameters were 0.04 and 0.02. Discussion on ascending aorta diameters and BA diameters has been addressed in the discussion section. One explanation for the lack of significant association may be related to the lack of power since mean ascending aorta diameter was 26.6 mm in IADE+ patients and 24.8 mm in IADE - (p=0.21). One other explanation is that the association does not exist, which could suggest that the pathophysiologic processes involved in IADE are different from the one involved in ascending aorta aneurysm (such as in Marfan�s disease). Other studies are needed to clarify this point.
References
1. Pico F, Labreuche J, Cohen AA, Touboul PJ, Amarenco P. Intracranial arterial dolichoectasia is associated with enlarged thoracic aorta. Neurology 2004 ;62 :2016-21
2. Pico F, Labreuche, Touboul P-J, Amarenco P. Intracranial arterial dolichoectasia and its association with atherosclerosis and stroke subtype. Neurology 2003;61:1736-1742.
3. Ubogu EE, Zaidat OO. Vertebrobasilar dolichoectasia diagnosed by magnetic resonance angiography and risk of stroke and death: a cohort study. J Neurol Neurosurg Psychiatry 2004;75:22-26
4. Smoker WR, Price MJ, Keyes WD, Corbett JJ, Gentry LR. High- resolution computed tomography of the basilar artery: 1. Normal size and position. Am.J.Neuroradiol. 1986;7:55-60
5. Hahn RT, Roman MJ, Mogtader AH, Devereux RB. Two-dimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol. 1989 ; 64 : 507-12
6. Lanza GM, Zabalgoitia-Reyes M, Frazin L et al. Plaque and structural characteristics of the descending thoracic aorta using transesophageal echocardiography. J Am Soc Echocardiogr. 1991 ; 4 : 19-28

Intracranial arterial dolichoectasia is associated with enlarged descending thoracic aorta 11 May 2005

Eroboghene E. Ubogu,
Department of Neurology, Louis Stokes Cleveland VAMC/ CASE School of Medicine
10701 East Blvd, Cleveland, OH 44106-1703,
Osama O. Zaidat
Send Correspondence to journal:
Re: Intracranial arterial dolichoectasia is associated with enlarged descending thoracic aorta

eroboghene.ubogu{at}case.edu Eroboghene E. Ubogu, et al.

The work of Pico et al describes an association between intracranial dolichoectasia (IADE) and enlarged descending thoracic aorta.[1] The authors suggest that this may provide evidence supporting dolichoectasia as a systemic arteriopathy. There are several study limitations that should be highlighted before generalizing these observations.
Measuring arterial flow-voids on axial T2-weighted magnetic resonance (MR) images, at pre-defined neuroanatomical landmarks does not establish the arterial length (�dolichos�), degree of deviation from shortest intracranial course (degree of tortousity) or true maximum diameter (�ectasia�), as ascertained with 3-dimension time of flight MR angiography for basilar and vertebral arteries.[2] Applying the published MR angiography criteria,[2] <15% of IADE stroke patients with basilar or vertebral artery involvement in the study of Pico et al would have been objectively diagnosed as �ectatic�.
The diagnostic gold standard for thoracic aorta disease is contrast angiography, although spiral computed tomography or MR angiography are replacing this clinically, as these are non-invasive with similar diagnostic accuracy.[3,4] Although transesophageal echocardiography (TEE) may be a valid measure, the authors do not establish its diagnostic sensitivity or specificity compared to contrast angiography.
Importantly, the authors do not define aortic enlargement. An increased aortic diameter does not necessarily imply dolichoectasia without normative values for aortic diameter, deviation from its shortest intra-thoracic course and length. The authors do not mention the anatomical locations of the ascending and descending aorta measured to establish diameter in their study, reducing independent validation of their methods.
The association between basilar artery diameter (as a marker of dolichoectasia) and aortic diameters is limited, as the basilar artery is only 1 of 11 large-vessel intracranial arteries encompassing IADE. The authors fail to provide the p-values for the association between basilar artery diameter and aortic diameters after co-variable adjustment, although they mention that these were significant. Assessing all 11 intracranial arteries would help establish a more accurate association between IADE and aortic diameter.
The authors mention that ascending aortic enlargement is a reliable marker for elastic tissue disease, but fail to adequately explain its lack of association with IADE in their study. Ascertaining dolichoectasia of the common carotid and vertebral arteries, as well as the ascending and descending aorta using angiographic techniques with well-defined diagnostic criteria and comparing these findings to IADE would be a better means of establishing association. If present, a potential systemic arteriopathy secondary to degeneration of the internal elastic lamina could be plausibly hypothesized.
References
1. Pico F, Labreuche J, Cohen A et al. Intracranial arterial dolichoectasia is associated with enlarged descending thoracic aorta. Neurology 2004;63:2016-2021.
2. Ubogu EE, Zaidat OO. Vertebrobasilar dolichoectasia diagnosed by magnetic angiography and risk of stroke and death: a cohort study. J Neurol Neurosurg Psychiatry 2004;75:22-26.
3. Krinsky G, Reuss PM. MR angiography of the thoracic aorta. Magn Reson Imaging Clin N Am 1998:6:293-320.
4. Fillinger MF. Imaging of the thoracic and thoracoabdominal aorta. Semin Vasc Surg 2000;13:247-263.
There are no conflicts of interest to report.