Abstract
Introduction
Information on the relationship between the diameters of the abdominal aorta (AA) and the basilar artery (BA) is limited. Using the Atahualpa Project cohort, we aimed to assess this association in Atahualpa residents aged ≥ 60 years.
Materials and methods
Atahualpa residents aged ≥ 60 years underwent abdominal ultrasound and MRA of intracranial vessels. The independent association between both arterial diameters was estimated in a generalized linear model adjusted for demographics, cardiovascular risk factors, the presence of a fetal-type Circle of Willis, and severity of white matter hyperintensities. Fractional polynomials were fitted to model the relationship between AA and BA diameters.
Results
The mean age of 277 participants was 69.5 ± 7.7 years (61% women). The mean AA diameter was 19.8 ± 3.3 mm, and the mean BA diameter was 3.1 ± 0.7 mm. The mean diameters of both arteries were significantly higher in men than in women. Locally weighted scatterplot smoothing showed a non-linear relationship between both arterial diameters. Fractional polynomial models showed that AA and BA diameters had a significant non-linear association in men (p = 0.005), but not in women (p = 0.315). When sex was excluded from a generalized linear model, the relationship between both arterial diameters became significant (p = 0.017).
Conclusion
In this population of community-dwelling older adults, the relationship between AA and BA diameters was confined to men. The finding of a large AA diameter in men should prompt the investigation of the intracranial vasculature because of the possibility of BA ectasia.
Keywords: Basilar artery, Abdominal aorta, Arterial diameters, Population-based study
Sommario
Introduzione
Esistono pochi dati relativi alla relazione tra i diametri dell’aorta addominale (AA) e dell’arteria basilare (BA). Utilizzando la coorte di pazienti del Progetto Atahualpa, è stata valutata questa associazione nei soggetti di Atahualpa di età ≥ 60 anni.
Materiale e metodi
I residenti di Atahualpa di età ≥ 60 anni sono stati sottoposti ad ecografia addominale e angiografia MR dei vasi intracranici. L’associazione indipendente tra entrambi i diametri arteriosi è stata stimata in un modello lineare generalizzato aggiustato per dati demografici, fattori di rischio cardiovascolare, presenza di un circolo di Willis di tipo fetale e gravità delle iperintensità della sostanza bianca. Polinomi frazionari sono stati adattati per modellare la relazione tra i diametri di AA e BA.
Risultati
L’età media dei 277 partecipanti era 69,5 ± 7,7 anni (61% donne). Il i diametro medio dell’AA era di 19,8 ± 3,3 mm e il diametro medio della BA di 3,1 ± 0,7 mm. Il diametro medio di entrambe le arterie era significativamente più alto negli uomini che nelle donne. Il livellamento dello scatter plot con ponderazione locale ha mostrato una relazione non lineare tra i diametri delle due arterie. I modelli polinomiali frazionari hanno dimostrato che i diametri AA e BA avevano un’associazione non lineare significativa negli uomini (p = 0005), ma non nelle donne (p = 0315). Quando il sesso è stato escluso dal modello lineare generalizzato, la relazione tra i due diametri arteriosi è risultata significativa (p = 0017).
Conclusione
In questa popolazione di adulti che vivono in comunità, la relazione tra i diametri di AA e BA è limitata agli uomini. Il ritrovamento di un grande diametro AA negli uomini dovrebbe sollecitare ad indagare il sistema vascolare intracranico a causa della possibilità di ectasia della BA.
Introduction
Basilar artery (BA) dolichoectasia is a non-atherosclerotic dilative arteriopathy associated with increased risk of stroke and other neurological complications [1]. This condition has been related to arterial wall thinning caused by degeneration of the internal elastic lamina and smooth muscular atrophy [2]. Other intracranial and extracranial arteries may also be affected by this process [3]. Anecdotal case reports, one autopsy study, and a clinical series conducted in stroke patients have suggested a link between BA dolichoectasia and dilatation of the descending aorta [4–6]. Nevertheless, available information on the aforementioned association may not be representative of the population at large. Using the Atahualpa Project cohort, we aimed to assess whether the BA’s diameter correlates with that of the abdominal aorta (AA) in community-dwelling older adults.
Materials and methods
Study population
Atahualpa is a rural village located in Coastal Ecuador where previous epidemiological studies have been conducted [7]. Residents are homogeneous regarding race (95% belong to the native/mestizo ethnic group), socio-economic status, and dietary habits; almost all the men are blue-collar workers, and most of the women are homemakers. These consistencies reduce the risk of unexpected confounders at the time of analyses.
Study design
Atahualpa residents aged ≥ 60 years identified during door-to-door surveys, who accepted the practice of MRI and had no contraindications for this procedure, were enrolled. Further consent was requested for the practice of an abdominal ultrasound. The independent association between AA and BA diameters was estimated after adjusting for relevant confounders (see below). The IRB of Hospital-Clínica Kennedy, Guayaquil, Ecuador (FWA 00006867) approved the study and the written informed consent forms.
Abdominal ultrasounds
These exams were performed using a Terason Smart 3300 (Teratech Corporation, Burlington, MA) and a 2–5 MHz convex probe. Following guidelines for assessment of the AA in adults [8], transverse images perpendicular to the long axis of the vessel were obtained below the diaphragm (near the celiac artery), at or near the level of the renal arteries, and above the iliac bifurcation. Measurements for analyses were taken at the greatest AA diameter (from the outer to outer edges) at any of these levels.
Neuroimaging protocol
Neuroimaging exams were performed with a Philips Intera 1.5T machine (Philips Medical Systems, Eindhoven), using previously defined protocols [9]. The BA’s diameter was measured using source MRA digital images at several points along the vertical course of the artery, and the maximal diameter was used for analyses [10]. Three-dimensional time-of-flight MRA sequences were used to identify individuals with a fetal-type Circle of Willis (CoW), defined when one P1 segment of the posterior cerebral artery was not visualized, and P2 and P3 segments were delineated continuously with the ipsilateral posterior communicating artery [11]. Using MRI, interest focused on the presence of white matter hyperintensities (WMH) of presumed vascular origin, defined as lesions appearing hyperintense on T2-weighted images that remained bright on FLAIR (without cavitation) and were graded according to the modified Fazekas scale [12]. Both the presence of a fetal-type CoW and the severity of WMH were used as confounders, since they have been related to the BA’s diameter in Atahualpa residents [10, 11].
Clinical covariates
Demographics and cardiovascular risk factors (smoking status, physical activity, diet, body mass index, blood pressure, fasting glucose, and total cholesterol blood levels) were assessed through interviews and procedures previously described in the Atahualpa Project [13].
Statistical analyses
Data analyses were carried out using STATA version 15 (College Station, TX). In univariate analyses, continuous variables were compared with linear models. Locally weighted scatterplot smoothing was used to evaluate the linearity of the association between AA and BA diameters. Then, we examined the independent association between both diameters in a generalized linear model adjusted for demographics, cardiovascular risk factors, the presence of a fetal-type CoW, and WMH severity. Fractional polynomials were fitted to model the relationship between AA and BA diameters.
Results
Of 437 individuals aged ≥ 60 years enrolled in the Atahualpa Project, 363 (83%) underwent a brain MRI and MRA of intracranial vessels. Five subjects were further excluded because their BAs had ≥ 50% focal stenosis or occlusion. Of the 358 remaining individuals, 277 (77%) accepted the practice of abdominal ultrasound and were included in the present study. The mean age of the participants was 69.5 ± 7.7 years (median age: 68 years), and 168 (61%) were women. A body mass index ≥ 30 kg/m2 was present in 58 (21%) persons, blood pressure ≥ 140/90 mmHg in 122 (44%), fasting glucose ≥ 126 mg/dL in 87 (31%), and total cholesterol levels ≥ 240 mg/dL in 40 (14%). Five (2%) participants were current smokers, 11 (4%) had a poor diet, and 20 (7%) had poor physical activity. Fifty-five (20%) individuals had moderate-to-severe WMH, and 77 (28%) had a fetal-type CoW.
The mean AA diameter was 19.8 ± 3.3 mm (median: 19.5 mm), and the mean BA diameter was 3.1 ± 0.7 mm (median: 3.1 mm). Univariate analyses showed several differences in the AA and BA diameters according to the covariates investigated (Table 1). The AA’s diameter was higher in men, in those with total cholesterol levels ≤ 239 mg/dL, and in those with a non-fetal-type CoW. The BA’s diameter was higher in people ≥ 69 years, in men, and in those with a non-fetal-type CoW. Locally weighted scatterplot smoothing showed a non-linear relationship between AA and BA diameters (Fig. 1). A fully-adjusted generalized linear model showed no association between AA and BA diameters (β 0.008; 95% C.I −0.017 to 0.033; p = 0.525); in this model, the covariates that remained significant were age (p = 0.030) and the presence of a fetal-type CoW (p < 0.001).
Table 1.
Mean (± SD) values of basilar artery and abdominal aorta diameters according to covariates included in this study (univariate analyses)
| Variable | Stratification | Basilar artery diameter | Abdominal aorta diameter | ||
|---|---|---|---|---|---|
| Mean ± SD (mm) | p value | Mean ± SD (mm) | p value | ||
| Age, median | ≤ 68 years (n = 144) | 2.95 ± 0.62 | 0.006* | 19.9 ± 2.84 | 0.508 |
| ≥ 69 years (n = 133) | 3.17 ± 0.71 | 19.64 ± 3.66 | |||
| Sex | Women (n = 168) | 2.99 ± 0.65 | 0.008* | 18.46 ± 2.52 | 0.001* |
| Men (n = 109) | 3.21 ± 0.69 | 21.81 ± 3.22 | |||
| Smoking status | Non- or past smokers (n = 272) | 3.07 ± 0.67 | 0.574 | 19.76 ± 3.28 | 0.635 |
| Current smokers (n = 5) | 3.24 ± 0.62 | 20.46 ± 1.99 | |||
| Body mass index | < 30 kg/m2 (n = 219) | 3.07 ± 0.7 | 0.614 | 19.79 ± 3.38 | 0.725 |
| ≥ 30 kg/m2 (n = 58) | 3.12 ± 0.54 | 19.73 ± 2.78 | |||
| Physical activity | Ideal/intermediate (n = 257) | 3.06 ± 0.67 | 0.225 | 19.77 ± 3.15 | 0.968 |
| Poor (n = 20) | 3.25 ± 0.72 | 19.8 ± 4.46 | |||
| Diet | Ideal/intermediate (n = 266) | 3.07 ± 0.66 | 0.268 | 19.7 ± 3.22 | 0.062 |
| Poor (n = 11) | 3.3 ± 0.96 | 21.57 ± 3.84 | |||
| Blood pressure | ≤ 139/89 mmHg (n = 155) | 3.05 ± 0.67 | 0.460 | 19.8 ± 3.28 | 0.899 |
| ≥ 140/90 mmHg (n = 122) | 3.11 ± 0.67 | 19.75 ± 3.24 | |||
| Fasting glucose | ≤ 125 mg/dL (n = 190) | 3.07 ± 0.67 | 0.731 | 19.71 ± 3.16 | 0.620 |
| ≥ 126 mg/dL (n = 87) | 3.1 ± 0.68 | 19.92 ± 3.48 | |||
| Total cholesterol blood levels | ≤ 239 mg/dL (n = 237) | 3.09 ± 0.67 | 0.339 | 19.98 ± 3.22 | 0.012* |
| ≥ 240 mg/dL (n = 40) | 2.98 ± 0.68 | 18.58 ± 3.27 | |||
| Circle of Willis | Non-fetal-type (n = 200) | 3.026 ± 0.62 | 0.001* | 20.02 ± 3.28 | 0.019* |
| Fetal-type (n = 77) | 2.59 ± 0.57 | 19.13 ± 3.14 | |||
| White matter hyperintensities | Non-to-mild (n = 222) | 3.04 ± 0.68 | 0.114 | 19.9 ± 3.12 | 0.290 |
| Moderate-to-severe (n = 55) | 3.2 ± 0.63 | 19.38 ± 3.75 | |||
*Statistically significant result
Fig. 1.
Locally weighted scatterplot smoothing showing a non-linear relationship between BA and AA diameters
Fractional polynomial models showed that AA and BA diameters had a significant non-linear association in men (p = 0.005), but not in women (p = 0.315). Among men, AA diameters between 15 and 25 mm had a positive association with BA diameter, while among women, there was no association. In view of these findings, sex was excluded from the previously constructed generalized linear model. In this new fitted model, the linear relationship between AA and BA diameters became significant (β 0.03; 95% C.I. 0.006–0.055; p = 0.017).
Discussion
The rationale for investigating a possible association between AA and BA diameters comes from the growing body of evidence suggesting that increases in both arteries’ sizes are related to an elastic tissue disorder and not to atherosclerosis [1–3]. In this population-based study, we found a non-linear association between AA and BA diameters, which was markedly influenced by sex. Men carried all the significance. This gender discrepancy is intriguing and possibly related to hormonal factors. We may also speculate that higher AA and BA diameters in men than in women make this association more evident in the former.
The clinical relevance of the findings of this study is evident, since most individuals with BA ectasia (or dolichoectasia) are initially asymptomatic, but the neurological complications related to these conditions may be catastrophic [1]. Therefore, the finding of a dilated AA during a routine abdominal sonogram (particularly in men) should prompt the investigation of the intracranial vasculature in search of an abnormally dilated BA.
A potential limitation of our study is the sonographic assessment of the AA, which might underestimate its diameter when compared with CT angiography [14]. On the other hand, the population-based design with unbiased enrollment of participants, together with the models used to assess sex-related differences in the relationship between BA and AA diameters, argue for the strength of our findings.
As previously noted, there are no other population-based studies attempting to correlate AA and BA diameters. Therefore, our results are not comparable with those of previous publications [4–6]. Further research is needed to assess the relationship between both arterial diameters and the pathogenic mechanisms implicated in this association.
Funding
Study supported by Universidad Espíritu Santo—Ecuador.
Compliance with ethical standards
Conflict of interest
The authors report that they have no conflict of interest to disclose.
Ethical standards
The study was conducted in accordance with the Helsinki Declaration. The Institutional Review Board of Hospital-Clínica Kennedy, Guayaquil (FWA 00006867) approved the study.
Informed consent
All participants signed the written informed consent form before enrollment.
References
- 1.Del Brutto VJ, Ortiz JG, Biller J. Intracranial artery dolichoectasia. Front Neurol. 2014 doi: 10.3389/fneur.2017.00344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pico F, Labreuche J, Amarenco P. Pathophysiology, presentation, prognosis, and management of intracranial arterial dolichoectasia. Lancet Neurol. 2015;14(8):833–845. doi: 10.1016/S1474-4422(15)00089-7. [DOI] [PubMed] [Google Scholar]
- 3.Gutierrez J, Bagci A, Gardener H, et al. Dolichoectasia diagnostic methods in a multi-ethnic, stroke-free cohort: results from the Northern Manhattan Study. J Neuroimaging. 2014;24(3):226–231. doi: 10.1111/j.1552-6569.2012.00781.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sacks JG, Linderburg R. Dolicho-ectatic intracranial arteries. Symptomatology and pathogenesis of arterial elongation and distention. Johns Hopkins Med J. 1969;125(2):95–106. [PubMed] [Google Scholar]
- 5.Gautier JC, Hauw JJ, Awada A, Loron P, Gray F, Juillard JB. Artères cérébrales dolichoectasiques. Association aux aneurysms de l’aorte abdominale. Rev Neurol (Paris) 1988;144(6–7):437–446. [PubMed] [Google Scholar]
- 6.Pico F, Labreuche J, Cohen A, Touboul PJ, Amarenco P, Investigators GENIC. Intracranial arterial dolichoectasia is associated with enlarged descending thoracic aorta. Neurology. 2004;63(11):2016–2021. doi: 10.1212/01.WNL.0000145845.12577.0F. [DOI] [PubMed] [Google Scholar]
- 7.Del Brutto OH, Mera RM, Castillo PR, Del Brutto VJ. Key findings from the Atahualpa Project: what should we learn? Expert Rev Neurother. 2018;18(1):5–8. doi: 10.1080/14737175.2018.1400382. [DOI] [PubMed] [Google Scholar]
- 8.Guideline developed in collaboration with the American College of Radiology; Society of Radiologists in Ultrasound AIUM practice guideline for the performance of diagnostic and screening ultrasound examinations of the abdominal aorta in adults. J Ultrasound Med. 2015;34(8):1–6. doi: 10.7863/ultra.34.8.15.13.0003. [DOI] [PubMed] [Google Scholar]
- 9.Del Brutto OH, Del Brutto VJ, Lama J. Sub-estudio de neuroimagen del Proyecto Atahualpa. Rev Ecuat Neurol. 2017;26(2):111–127. [Google Scholar]
- 10.Del Brutto OH, Mera RM, Del Brutto VJ, Costa AF, Zambrano M, Brorson J. Basilar artery dolichoectasia: prevalence and correlates with markers of cerebral small vessel disease in community-dwelling older adults. J Stroke Cerebrovasc Dis. 2017;26(12):2909–2914. doi: 10.1016/j.jstrokecerebrovasdis.2017.07.014. [DOI] [PubMed] [Google Scholar]
- 11.Del Brutto OH, Mera RM, Costa AF, Del Brutto VJ. Basilar artery diameter is inversely associated with fetal type circle of Willis. Eur Neurol. 2017;78(3–4):217–220. doi: 10.1159/000480430. [DOI] [PubMed] [Google Scholar]
- 12.Pantoni L, Basile AM, Pracucci G, et al. Impact of age-related cerebral white matter changes on the transition to disability—the LADIS study: rationale, design and methodology. Neuroepidemiology. 2005;24(1–2):51–62. doi: 10.1159/000081050. [DOI] [PubMed] [Google Scholar]
- 13.Del Brutto OH, Santamaría M, Ochoa E, Peñaherrera E, Santibáñez R, Pow-Chon-Long F, Zambrano M, Del Brutto VJ. Population-based study of cardiovascular health in Atahualpa, a rural village of coastal Ecuador. Int J Cardiol. 2013;168(2):1618–1620. doi: 10.1016/j.ijcard.2013.01.017. [DOI] [PubMed] [Google Scholar]
- 14.Manning BJ, Kristmundsson T, Sonesson B, Resch T. Abdominal aortic aneurysm diameter: a comparison of ultrasound measurements with those from standard and three-dimensional computed tomography reconstruction. J Vasc Surg. 2009;50(2):263–268. doi: 10.1016/j.jvs.2009.02.243. [DOI] [PubMed] [Google Scholar]