Multi-perspective ultrasound evaluation of Carotid Intima-Media thickness in hemodialysis access patients and its possible use to predict clinical outcomes

Abstract

Background

Hemodialysis (HD) patients present an increased carotid intima-media thickness (CIMT) and calcified plaques, and studies have demonstrated that CIMT is predictor for cardiovascular death in this patient population; hence, the importance of identifying early non-invasive markers of atherosclerosis. We aim to propose a new three-perspective CIMT measuring method in HD population, and to further investigate the possible use and value of this method to predict outcomes.

Methods

A cross-sectional study was conducted, the CIMT was measured with duplex ultrasound during the perioperative access planning. Software provided maximum CIMT in a 10 mm (mm) width of the common carotid artery. CIMT was measured in Right (R) and Left (L) carotid arteries, both in anterior (a), lateral (l), and posterior (p) view. The sum of these values (Ra+Rl+Rp+La+Ll+Lp) was completed and termed as Sixth Carotid Intimal (SCI) score. We stratified either in low (summation <4) or high (>=4) SCI score groups. Demographics, patency rates, complications, and mortality were collected; chi-square test was used for our analysis.

Results

A total of 30 patients (mean age 49 years and 56% females) that underwent access creation in the upper arm from 2010 to 2016 were selected and studied. A low SCI score was found 19 patients and 11 had high SCI, no significant difference was observed in demographics, history cardiovascular disease, and clinical outcomes as early thrombosis, and primary, primary-assisted patency at 12 months. Interestingly, during the follow up period of 36 months a significant higher difference in late thrombosis rates occurred (63.6% vs 26.3%, p = .044) and a higher all-cause mortality (54.4% vs 15.7%, p = .025) in patients with SCI score group above 4.

Conclusions

The SCI score method, might offer a screening tool for traditional cardiovascular risk factors in HD patients. In this study, we demonstrate an increased rate in late thrombosis and mortality in those with high SCI. Further research is necessary to better define the role of CIMT in vascular surgical procedures.

Keywords

Carotid intima-media thickness hemodialysis access arteriovenous fistula vascular access outcomes multi-perspective carotid intima-media thickness measurement

Introduction

Arteriovenous fistula (AVF) is considered the preferred vascular access in patients with end-stage renal disease (ESRD) in need for renal replacement therapy with hemodialysis (HD).¹ Even when the objective of this autologous option is to constitute a long-term access, the blood vessels regain the physiological levels of wall shear stress (WSS) after AVF creation,^1,2 and these may attempt to decrease WSS due to the increased blood flow secondary to vascular dilation.^2,3 Atherosclerosis, also affects multiple vascular beds, including those in the upper extremity and may have an impact in AVF maturation.^1–3

End-stage renal disease (ESRD) is recognized as an independent cardiovascular risk factor, accounting for a relative risk of 1.4 to 3.2 for coronary artery disease (CAD) and 2.3 to 3.5 for stroke.⁴ HD patients suffers from accelerated atherosclerotic disease, and they manifest an increased intima-media thickness (IMT) and more intensely calcified plaques in comparison with the atherosclerotic lesions found in the general population.^5–7 As a result of these factors, this population is 5–30 times more likely to die from cardiovascular events. Nowadays, intima-media thickness (IMT) measurement widely applies to stratifying cardiovascular risk in primary prevention.⁷ Indeed, it is a valuable predictor of subsequent CAD and stroke, the leading causes of cardiovascular death in the HD population; ^7–9 hence, the importance of developing early noninvasive markers of atherosclerosis.^8,9 This lead us to investigate the impact of IMT in clinical outcomes in this complex group of patients. Notably, in the common carotid artery (CCA) there is a distinctive biological behavior between the atherosclerotic plaque and the carotid IMT (CIMT); of the two methods described in the literature for measurement of CIMT, for the purpose of our study, we decided to exclude the plaque method to avoid the inclusion of patients with advanced arterial disease, this method allows the description of changes related to the vessel wall thickness,¹⁰ vascular remodeling and early atherosclerosis.^8,11 Additionally, the outcome predictions seem to be more precise when measuring maximal CIMT compared to the inclusion of the atherosclerotic plaque during CCA evaluation.^12,13 According to the literature, the latter is higher associated with CAD.^8,14,15 Moreover, IMT measurement accuracy is dependent on slight variations in the common carotid artery imaging location, the type of imaging device, the measurement process adopted, as well as the number of images acquired.¹⁶ Thus, we thought of the possibility of improving this single measurement method, by adding the maximal IMT values obtained from 3 perspectives of the artery. This addition may allow us to determine more precisely the IMT global state by measuring it in a three-dimensional view. Furthermore, we considered using the values obtained from both carotid arteries due to the systemic and symmetric nature of the atherosclerotic disease.¹⁷

Our main objective was to propose a novel method for measuring CIMT in hemodialysis patients; consequently, to determine if this method derived from the CCA CIMT biology may be a valuable parameter to study and compare and possibly predict outcomes in patients who undergo upper arm arteriovenous access creation for hemodialysis access in terms of patency, complications, and mortality rates.

Methods

We performed a cross-sectional study during the preoperative period of upper arm arteriovenous access creation; this study was conducted under approval by the Institutional Review Board, and informed consent was obtained from all patients. We selected 30 subjects with chronic kidney disease who underwent these procedures from 2010 to 2016, the minimum follow-up was 24 months, patients were selected if they had either an autologous brachiocephalic AVF or an arteriovenous graft (AVG) between the brachial artery and the basilic vein. Autologous AVFs were created through a direct non-transposed end-to-side anastomosis between the brachial artery and the cephalic vein whenever the pre-operative ultrasound vascular diameters were suitable, meaning greater than 3 mm (mm) and a non-significant difference in Allen’s test nor the inter-arm systolic blood pressure, was found. If criteria were not fulfilled, we conducted an arteriovenous graft with the aid of a 6 mm three-layered expanded Polytetrafluoroethylene (ePTFE) graft. As part of the preoperative assessment, contrast venography confirmed superficial and deep vein system patency since many of these patients had already undergone radiocephalic fistula, had a history of multiple previous temporary HD catheter utilization, or both. Demographics, comorbidities and history and relevant information related to arteriovenous access planning, function, patency, complications, and mortality were collected.

Measurement of CCA intimal thickness

Our protocol included an automated measurement of CIMT accomplished with the SonoCalc 5.0 IMT software of SonoSite Edge Ultrasound FUJIFILM Inc. The exam was performed with a 10 megahertz (Hz) linear array probe, started from the proximal CCA towards the bifurcation, followed by the internal and external carotid arteries. Measurements were done one cm below the carotid bifurcation on the CCA. Between the lines shown in Figure 1, the software processed the values by an algorithm that determined mean and maximal IMT commonly in a 10 mm width of CCA. Gain controls were allowed to be regulated by the sonographer, without a specific control setting, to obtain the best possible picture. In Figure 1, details of a graphical description of measurements completed one cm below the carotid bifurcation on the CCA. The IMT images selected in the B mode image were from the deeper wall (far from the transducer) as they have constant and repeatable visualization, yielding better imaging and thus supporting its employment for our measurements, therefore, our conclusions can only apply to deeper walls. Longitudinal views were taken first from the left CCA followed by the right CCA; we examined each CCA in the anterior, lateral, and posterior positions. Measurements were recorded for each subject on the same day by one sonographer, a certified physician by the Mexican Council and Board of Angiology, Vascular, and Endovascular Surgery who completed the CIMT measurements with a single ultrasound system.

Figure 1.

Graphical description of the measurements performed one centimeter below the carotid bifurcation on the CCA. The IMT images selected in the B mode image were from the deeper wall (far from the transducer) as they have constant and repeatable visualization, yielding better imaging and thus supporting its employment for our measurements.

Sixth Carotid Intimal (SCI) Score

After CIMT was measured and labeled in the Right (R) and Left (L) CCA, both arteries were visualized in anterior (a), lateral (l), and posterior (p) view. The score comprised values exclusively from the maximal CIMT of this measurement on behalf of the vascular biology we intended to study in this research. The sum of these six measures (Ra+Rl+Rp+La+Ll+Lp), and afterward, calculation of a mean value. We designated and termed the resulting value as the Sixth Carotid Intimal (SCI) score. (Figure 2) Then, based on a previous study where researches measured CIMT in patients without cardiovascular risk factors and reported a maximum thickness in 1.11 mm in men and 1.13 mm in women,¹⁸ we decided to divide our population into two groups: low (<4 mm) or high (>=4 mm) SCI scores. Figure 2 is a graphical description of the Sixth Carotid Intimal Score-making process. Notably, when taking into account the difference in IMT due to the change in diameter between diastole and systole, averaged to 0.041 mm (95% CI: 0.039 to 0.042 mm), all patients remained in the same SCI score group.¹⁹ Nonparametric statically analyses were used for comparison with a p value <.05 as significant.

Figure 2.

Graphical description of the Sixth Carotid Intimal Score-making process. Notably, when taking into account the difference in IMT due to the change in diameter between diastole and systole, averaged to 0.041 mm (95% CI: 0.039 to 0.042 mm), all patients remained in the same SCI score group.

Follow-up was performed in the outpatient clinic with ultrasound, and patients were cannulated for the first time after 6 weeks after the surgery; a greater than 50% diameter reduction of the vessel (or graft) associated with one or more hemodynamic, functional, or clinical features not explained by other reasons was defined as hemodynamically significant stenosis. Clinical manifestations found were: limb swelling, prolonged bleeding after HD sessions, abnormal pulsatile fistula or graft and diminished thrill. Early thrombosis was considered when the access did not develop or within the first 3 months of its use.²⁰ The same period was considered for early infection criteria. The time elapsed between AVF creation until we did first intervention (endovascular or open surgical) to maintain or restore blood flow is defined as unassisted primary patency. Whenever an intervention, usually endovascular, was needed for improving fistula maturation or performance until the next access thrombosis or reintervention, established primary-assisted patency. The time elapsed between AVF creation and abandonment or achievement of a censored event (irreparable dysfunction or re-thrombosis) with or without the intervention for recanalization, defined as secondary patency.²¹ Lastly, we also analyzed the association between AVF outcomes and SCI score with the Chi-square test, considering p to be significant when <0.05.

Results

From all the upper-arm AVFs included, twenty-eight patients (93.3%) underwent an autologous brachiocephalic fistula and two brachiobasilic arteriovenous grafts procedure averaged 15 mm of complete blood loss. The mean age of patients was 51.3 years in the high SCI score, compared to 46.8 years in the low SCI score. Gender distribution demonstrated no differences between groups, corresponding approximately half of the population to women. Comorbidities such as diabetes mellitus, arterial hypertension, smoking, hypercoagulable disorders, heart disease, and stroke were similarly distributed in both SCI groups; the etiology of ESRD showed no statistical difference between groups. It is important to note that diabetic and idiopathic nephropathies were the most frequent etiologies encountered. Distribution was even for pharmacological therapy, such as antiplatelet and anticoagulation therapy. Demographics description and distribution between groups are detailed in Table 1. Eighty percent of the subjects had the AVF created in the left arm. Two-thirds of patients had placed at least one hemodialysis catheter before AVF surgery (90% in high SCI group vs 78.9% in low SCI group), and the right internal jugular was the most common vessel. Table 2 provides details regarding the preoperative AVF planning. Both groups had a distribution of arteriosclerotic plaque presence equitably, although its incidence tended to be higher in high SCI score group (36.3% vs 10.5% in low SCI group). The plaque found in this study was non-obstructive and located in the CCA (away from the score measurement zone concerned). The plaque was not statistically associated with CAD, accounting for 2 cases in the high SCI group and one in the low SCI group. On the other hand, two patients suffered from a stroke during the follow-up period, which was not related to atherosclerotic plaque presence or high SCI score (Table 2).

Table 1.

Patient’s demographics and comorbidities.

	SCI >=4 (N = 11)	SCI <4 (N = 19)	p value
Age (mean)	51.3 (33-84)	46.8 (22-78)
Female	6 (54.5%)	11 (57.8%)	0.858
Comorbidities
Diabetes	6 (54.5%)	8 (42.1%)	0.510
Hypertension	6 (54.5%)	11 (57.8%)	0.858
Smoking	2 (18.1%)	2 (10.5%)	0.552
Thrombophilia	1 (9.0%)	3 (15.7%)	0.602
CAD	2 (18.1%)	1 (5.2%)	0.255
Stroke	0 (0%)	2 (10.5%)	—
Etiology ESRD
Diabetic	5 (45.5%)	7 (36.8%)	0.642
Idiopathic	4 (36.6%)	2 (10.5%)	0.088
Lupus	1 (9.0%)	3 (15.7%)	0.602
Drugs
ASA	1 (9.0%)	1 (5.2%)	0.685
Anticoagulation	0 (0%)	4 (21.0%)	—

*SCI. Sixth Carotid Intimal Score, CAD. Coronary artery disease, ESRD. End-Stage Renal Disease, ASA. Acetylsalicylic Acid.

Table 2.

Arterio-venous fistula planning.

	SCI >=4 (N = 11)	SCI <4 (N = 19)	p value
Access history
Previous catheter	10 (90%)	15 (78.9%)	0.396
>2 catheter	3 (27.2%)	5 (26.4%)	0.954
RIJ catheter	8 (72.7%)	11 (57.8%)	0.416
LIJ catheter	4 (36.3%)	6 (31.5%)	0.788
RS catheter	4 (36.3%)	5 (26.4%)	0.562
LS catheter	5 (45.4%)	2 (10.5%)	0.029
Previous AVF	4 (36.3%)	0	—
AVF location
Upper arm	10 (90.9%)	19 (100%)	—
Dominant	2 (18.1%)	5 (26.4%)	0.611
AVF type
Autologous^b	9 (81.8%)	14 (73.6%)	0.611
Atherosclerotic plaque^c	4 (36.3%)	2 (10.5%)	0.088
CVD	2 (50%)	1 (50%)	1000

^aAVF. Arteriovenous fistula, SCI. Sixth Carotid Intimal Score, RIJ Right internal jugular, LIJ. Left internal Jugular, RS. Right subclavian, LS. Left subclavian, CVD. Cardiovascular disease.

^bComparison between autologous vs graft AVF.

^cAtherosclerotic plaque in the CCA out from the zone considered for SCI score measurement.

Most fistulas maintained acceptable patency throughout the analysis, the primary patency was 81% for high SCI score compared to 52.6% for low SCI score. During the follow up period only 5 AVFs required an interventional procedure to improve fistula performance. Conversely, we examined surgically 4 AVFs to regain patency. Primary-assisted patency corresponded to 100% for high SCI score versus 89% for low SCI score during the lapse recorded of 12 months. Primary nor primary assisted patency did not show significant differences. Complications such as early thrombosis, and early and late infection were similar among both groups; late thrombosis showed a significant difference when analyzed, favoring the high SCI score group (63.6% high SCI group vs 26.3% low SCI group, p = .044) at 36 months. Remarkably, the analysis demonstrated the same proclivity in the high SCI score group for mortality (54.5% high SCI group vs 15.7% low SCI group, p = .025) during the 36 months of follow up (Table 3).

Table 3.

Analysis of clinical outcomes.

	SCI >=4 (N = 11)	SCI <4 (N = 19)	p value
Patency rates
Primary^a	9 (81.8%)	13 (68.4%)	0.423
Assisted^a	2 (18.1%)	6 (31.5%)	0.423
Complications
Early infection	1 (9.09%)	1 (5.26%)	0.685
Late infection	0	1 (5.26%)	—
Early thrombosis	1 (9.09%)	1 (5.26%)	0.685
Late thrombosis^b	7 (63.6%)	5 (26.3%)	0.044
Others
Mortality^b (all cause)	6 (54.5%)	3 (15.7%)	0.025

SCI Sixth Carotid Intimal Score.

^aPrimary and Assisted primary at 12 months.

^bMortality and late thrombosis follow up was at 36 months.

Discussion

The sixth Carotid Intimal score measurement of the deeper wall that we propose in this paper is technically feasible with non-invasive imaging techniques. This score might offer some advantages over the single measurement of CIMT, like being precise in the early stages of atherosclerotic disease and possibly a more accurate CIMT value due to the multiple images obtained from a three-perspective view. This comprehensive assessment may allow detecting arterial wall changes in areas without identified plaques that neither Doppler ultrasound, computed tomographic angiography, nor magnetic resonance can give more details in the early stages of atherosclerotic disease.²² As we propose the SCI score for the first time with this article, further studies and validation are necessary.

Our findings differ from descriptions in terms of isolated CIMT measurements; in our population, almost all CIMT were calculated below 0.9 mm and therefore evaluated as normal (range 0.1 to 1.12 mm, the mean value of 0.6 mm). These findings seem to be uncommon for HD patients, largely known to exhibit early atherosclerosis, and contrast to other publications where mean maximal CIMT measurements were 3.4 mm (range 1.3 to 5.6 mm) in this patients’ population.⁹ The well-described difference between end-diastolic and peak systolic CIMT measurement (increased with pulse pressure) may not explain these results as patients remained in the same SCI group despite these variations, and possibly ethnicity could explain these findings.¹⁹ Another interesting factor and contributing aspect, is the fact that most CIMT measurements in previous studies were obtained from middle-aged or older subjects, and scarce studies have been done to study CIMT HD in relatively young patients as the present study with a mean age of 49 years.

We must take into account the different factors involve in the pathogenesis in each vascular bed, it is now understood that several ultrasonographic phenotypes of atherosclerosis, such as IMT, plaque area, and stenosis are biologically and genetically distinct.^23–26 Future analysis should be focused on clarifying these markers’ associations in terms of vascular bed location and biological behavior. Additionally, multiple previous HD catheter placements may increase the abnormal flow conditions after AVF creation.²⁷ The fistula remodeling response remains to be elucidated in this group of patients.

The cardiovascular risk might also be assessed in HD patients by the SCI score measurement in patients with an upper-arm access. The previous descriptions by Lorenz et al.²⁶ showed that a CIMT increment of 0.1 mm increased the risk of myocardial infarction from 10% to 15% and the risk of stroke from 13% to 18% after adjusting for age and sex, this has also supported this hypothesis.⁴ Even though the atherosclerotic plaque presence did not prove a significant association, the high SCI score group showed a clear tendency that could result in a potentially significant difference when increasing the sample size.²⁷ Likewise, the subanalysis of plaque presence pooling HD population (high and low SCI group), demonstrated congruency with other publications, favoring the association of CCA atherosclerotic plaque and CAD (X2 = 8.72 p = .003) (Table 4). The results suggest that CCA atherosclerotic finding in HD patients with upper arm AVF is still a marker of CAD but do not support a role for high SCI score in any cardiovascular event; despite this fact, a high SCI score did prove significant in terms of all-cause mortality for HD patients in the perioperative period of an upper arm AVF (54.5% vs 15.7%, p = .025). Other non-invasive atherosclerotic indexes, such as brachial-ankle pulses wave velocity and cardio-ankle vascular index (CAVI), have been equally evaluated in HD patients to determine their arterial stiffness; ^28,29 the former is similarly able to predict increases in maximal CIMT. Although changes in water volume before and after HD influence CAVI, these non-invasive tests were significantly higher in HD patients than the age and gender-matched controls, independent from age and systolic blood pressure.³⁰

Table 4.

CCA atherosclerotic plaque and CVD in HD patients.

	CVD +	CVD −	n
Atherosclerotic plaque +	3 (0.80) [6.05]^a	3 (5.20) [0.93]	6
Atherosclerotic plaque −	1 (3.20) [1.51]	23 (20.80) [0.23]	24
Number	4	26	N = 30

CCA Common Carotid artery. CVD Cardiovascular disease. HD, Hemodialysis.

^aThe chi-square statistic is 8.726. The p-value is 0.003137. The result is significant at p < 0.05.

Once the arteriovenous circuit is created, vessels tend to develop extensive intimal remodeling as a protective response to regain the physiological WSS; ^2,3 thus, this intimal remodeling is known as neointimal hyperplasia (NH) which is the consequence of the increased flow rates that elevates blood velocity into the venous segment that results in a high WSS.² This overwhelming reaction impacts access function and increases the risk of failure.^2,3 Surveillance of vascular access assures durable and well-functioning access; it intends to identify stenotic lesions and their consequent undesirable thrombosis; thus, it focuses on determining treatable factors to prevent AVF failure. Due to the complexity of the multiple elements involved in vascular access failure, many screening methods for hemodynamic performance have been proposed, mostly related to the parameters during the hemodialysis session, in addition to clinical surveillance, other methods include dilution techniques based on different blood properties,^31–35 transcutaneous access flow,³⁴ and the variable flow Doppler.³⁵ The addition of the SCI score to these measurements may assist to predict cardiovascular risk and vascular access patency rates in this group of patients, further research is warranted. Radio-opaque markers, casting, and micromagnetic resonance imaging are potential tools for establishing the link between the history of hemodynamics involved in the AVF remodeling process and the changes in luminal diameter and IMT.³⁶ Unfortunately, these tools may also be limited by their availability and invasiveness. Thus, the statistical association between the High SCI score and late thrombosis of AVF in HD patients may pose the SCI score as a surveillance tool, rendering it possible to identify AVFs at risk of late thrombosis. The main pathophysiologic change of the radial artery associated with the early failure of AVF in HD patients is the NH in the radial artery,³⁷ the SCI score association with radial artery IMT to predict NH in this vessel before the AVF creation remains unclear, and this import subject needs to be addressed and carefully studied in this complex population through well design clinical studies. Most of the patients in our manuscript were seen and treated initially in outside institutions, before our evaluation for access planning, preventing us from avoiding the utilization of multiple hemodialysis catheters before performing a fistula, and this fact limited us to accurately determine the etiology of ESRD explaining the great number of patients with idiopathic disease. Even when this confounding factor was evenly distributed in both groups, the majority of left subclavian (LS) catheters placed were in patients of the high SCI group. This catheter location is a well-known factor for central venous stenosis, also considered a disadvantage for long-term patency in AVF. Other important factor, is the possibility of stenotic lesions or complete occlusion of the ICA, although none of the patients in the present series had carotid disease, the presence of these could impact in CIMT measurement and should be in consideration.

Besides the known limitations of cross-sectional designs and retrospective analysis, this is the first study in which the SCI score, a non-invasive three-perspective CIMT measurement, has been proposed, which may also be a useful and practical tool for determining late thrombosis and mortality in HD patients with an upper-arm arteriovenous access. Although this is a small number, our initials results justify score validation and further analysis in a larger sample population.

Conclusion

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Adriana Torres-Machorro

Carlos A. Hinojosa

Appendix

References

Asif

Roy-Chaudhury

Beathard

. Early arteriovenous fistula failure: a logical proposal for when and how to intervene. Clin J Am Soc Nephrol 2006; 1(2): 332–339.

Vita

Holbrook

Palmisano

, et al. Flow-induced arterial remodeling relates to endothelial function in the human forearm. Circulation 2008; 117(24): 3126–3133.

Longenecker

Coresh

Powe

, et al. Traditional cardiovascular disease risk factors in dialysis patients compared with the general population: the CHOICE Study. J Am Soc Nephrol 2002; 13(7): 1918–1927.

Simon

Megnien

Chironi

. The value of carotid intima-media thickness for predicting cardiovascular risk. Arterioscler Thromb Vasc Biol 2010; 30(2): 182–185.

von Baeyer

Hopfenmüller

Riedel

, et al. Atherosclerosis: current concepts of pathophysiology and pharmacological intervention based on trial outcomes. Clin Nephrol 2003; 60(Suppl 1): S31–S48.

Ross

. Atherosclerosis — An inflammatory disease. N Engl J Med 1999; 340(2): 115–126.

Nassiri

Hakemi

Asadzadeh

, et al. Differences in cardiovascular disease risk factors associated with maximum and mean carotid intima-media thickness among hemodialysis patients. Iran J Kidney Dis 2012; 6(3): 203–208.

Simon

Gariepy

Chironi

, et al. Intima-media thickness: a new tool for diagnosis and treatment of cardiovascular risk. J Hypertens 2002; 20(2): 159–169.

Krasniak

Drozdz

Pasowicz

, et al. Factors involved in vascular calcification and atherosclerosis in maintenance hemodialysis patients. Nephrol Dial Transplant 2006; 22(2): 515–521.

10.

Polak

Szklo

Kronmal

, et al. The value of carotid artery plaque and intima‐media thickness for incident cardiovascular disease: the multi‐ethnic study of atherosclerosis. J Am Heart Assoc 2013; 2(2): e000087.

11.

Salonen

. Ultrasonographically assessed carotid morphology and the risk of coronary heart disease. Arterioscler Thromb 1991; 11(5): 1245–1249.

12.

Prati

Tosetto

Vanuzzo

, et al. Carotid intima media thickness and plaques can predict the occurrence of ischemic cerebrovascular events. Stroke 2008; 39(9): 2470–2476.

13.

Ebrahim

Papacosta

Whincup

, et al. Carotid plaque, intima media thickness, cardiovascular risk factors, and prevalent cardiovascular disease in men and women: the British Regional Heart Study. Stroke 1999; 30(4): 841–850.

14.

Desikan

Crichton

Hoang

, et al. Effect of exhaust- and nonexhaust-related components of particulate matter on long-term survival after stroke. Stroke 2016; 47(12): 2916–2922.

15.

Chan

Mancini

Kuramoto

, et al. The prognostic importance of endothelial dysfunction and carotid atheromaburden in patients with coronary artery disease. J Am Coll Cardiol 2003; 42(6): 1037–1043.

16.

Vartiainen

Laatikainen

Peltonen

, et al. Predicting coronary heart disease and stroke: the FINRISK calculator. Glob Heart 2016; 11(2): 213–216.

17.

Adams

Simoni

Bordelon

Jr , et al. Bilateral symmetry of human carotid artery atherosclerosis. Stroke 2002; 33(11): 2575–2580.

18.

Jarauta

Mateo-Gallego

Bea

, et al. Carotid intima-media thickness in subjects with no cardiovascular risk factors. Rev Esp Cardiol 2010; 63(1): 97–102.

19.

Polak

Johnson

Harrington

, et al. Changes in carotid intima‐media thickness during the cardiac cycle: the multi‐ethnic study of atherosclerosis. J Am Heart Assoc 2012; 1(4): e001420.

20.

Beathard

Arnold

Jackson

, et al. Aggressive treatment of early fistula failure. Kidney Int 2003; 64(4): 1487–1494.

21.

Lee

Mokrzycki

Moist

, et al. Standardized definitions for hemodialysis vascular access. Semin Dial 2011; 24(5): 515–524.

22.

Pignoli

Tremoli

Poli

, et al. Intimal plus medial thickness of the arterial wall: a direct measurement with ultrasound imaging. Circulation 1986; 74(6): 1399–1406.

23.

Spence

Hegele

. Noninvasive phenotypes of atherosclerosis. Arterioscler Thromb Vasc Biol 2004; 24(11): e188–e189.

24.

Barth

. Carotid intima media thickness and beyond. Curr Drug Targets - Cardiovasc Hematol Disord 2004; 4: 129–145.

25.

Pollex

Spence

House

, et al. A comparison of ultrasound measurements to assess carotid atherosclerosis development in subjects with and without type 2 diabetes. Cardiovasc Ultrasound 2005; 3(1): 15.

26.

Lorenz

Markus

Bots

, et al. Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis. Circulation 2007; 115(4): 459–467.

27.

Duque

Tabbara

Martinez

, et al. Dialysis arteriovenous fistula failure and angioplasty: intimal hyperplasia and other causes of access failure. American Journal of Kidney Diseases. 2017;69(1):147–151. Faber J, Fonseca LM. How sample size influences research outcomes. Dental Press Journal of Orthodontics. 2014;19(4):27–9.

28.

Yambe

Yoshizawa

Saijo

, et al. Brachio-ankle pulse wave velocity and cardio-ankle vascular index (CAVI). Biomed Pharmacother 2004; 58: S95–S98.

29.

Kubozono

Miyata

Ueyama

, et al. Clinical significance and reproducibility of new arterial distensibility index. Circ J 2007; 71(1): 89–94.

30.

Ueyama

Miyata

Kubozono

, et al. Noninvasive indices of arterial stiffness in hemodialysis patients. Hypertens Res 2009; 32(8): 716–720.

31.

Gotch

FA1

Buyaki

Panlilio

, et al. Measurement of blood access flow rate during hemodialysis from conductivity dialysance. American Society for Artificial Internal Organs Journal 1999; 45(3): 139–146.

32.

Jiang

Clayton

Maguire

, et al. Validation of the measurement of haemodialysis access flow using a haemoglobin dilution test. Blood Purif 2011; 32(1): 48–52.

33.

Schneditz

Wang

Levin

. Validation of haemodialysis recirculation and access blood flow measured by thermodilution. Transplantation 1999; 14(2): 376–383.

34.

Steuer

Miller

Zhang

, et al. Noninvasive transcutaneous determination of access blood flow rate. Kidney Int 2001; 60(1): 284–291.

35.

Weitzel

Rubin

Leavey

, et al. Analysis of variable flow Doppler hemodialysis access flow measurements and comparison with ultrasound dilution. Am J Kidney Dis 2001; 38(5): 935–940.

36.

Rajabi-Jagahrgh

Roy-Chaudhury

Wang

, et al. New techniques for determining the longitudinal effects of local hemodynamics on the intima-media thickness in arteriovenous fistulae in an animal model. Semin Dial 2013; 27(4): 424–435.

37.

Kim

Song

Yoon

, et al. Preexisting intimal hyperplasia of radial artery is associated with early failure of radiocephalic arteriovenous fistula in hemodialysis patients. Am J Kidney Dis 2003; 41(2): 422–428.