Introduction

Arteriosclerosis is an important risk factor for cardiovascular events in patients with chronic kidney disease (CKD). Cardiovascular disease is the leading cause of death in patients with end-stage kidney disease (ESKD). In fact, the cardiovascular mortality rate in hemodialysis (HD) patients is 10–20 times higher than in the general population1. This heightened risk is closely related to coronary artery disease and valvular heart disease2.

Aortic stenosis (AS) is one of the most common valvular diseases worldwide. In HD patients, the prevalence of aortic valve calcific abnormalities ranges from 28 to 85%, with severe AS detected in 6–13% of cases3. Additionally, the prevalence of AS and mitral regurgitation is higher in patients with CKD, including those not on dialysis4. ESKD is also associated with aortic and mitral valve calcification, accelerating the progression of valve stenosis and leading to worse outcomes5. Factors such as CKD-mineral and bone disorder, inflammation, anemia, hemodynamic changes, and volume overload contribute to the progression of AS3,6. The identification of valvular calcification on echocardiography has been reported to be independently associated with adverse cardiovascular outcomes7. Aortic valve calcification increases the risk of cardiovascular mortality, even in the absence of significant obstruction of left ventricular outflow8. In patients with HD, calcified AS has also been identified as an independent risk factor for mortality9,10,11.

Patients with hemodynamically significant AS are vulnerable to episodes of profound hypotension on HD10, a condition known as dysdialysis syndrome, which can be life-threatening. In the same context, AS poses a significant challenge in managing patients on HD. AS progression plays a crucial role in the prognosis of patients with ESKD. However, there is limited research on the factors contributing to AS progression in dialysis patients, and no reports have explored the association between histological findings in arterial tissue and AS progression. Because human arterial tissue can be easily harvested during arteriovenous fistula (AVF) surgery in patients with HD, we focused on the associations of arterial calcification and changes in peak flow velocity (ΔVmax) with AS progression. According to clinical guidelines, peak aortic jet velocity (Vmax) is also used in the diagnostic criteria and severity classification of AS12. Generally, an increase in Vmax indicates the progression of stenotic changes of aortic valve, even before AS manifests clinically. Reportedly, Vmax gradually rises from the early stages of aortic valve sclerosis and accelerates over time, ultimately leading to the development of AS13.

As previously mentioned, CKD is associated with arteriosclerosis and AS. Hypothetically, patients with ESKD and arterial calcification are at a higher risk of developing AS.

In this study, we aimed to identify factors contributing to the progression of stenotic changes of aortic valve in patients with HD, including histological changes in peripheral arteries harvested at the time of AVF creation.

Results

Baseline characteristics

The clinical and laboratory baseline characteristics of the enrolled patients are presented in Table 1. The mean age at the time of AVF creation was 67.3 ± 13.6 years, with 37.2% of the participants (n = 35) being females. All patients had ESKD, with an average serum creatinine level of 8.46 ± 2.91 mg/dL and an estimated glomerular filtration rate of 5.7 ± 1.9 mL/min/1.73 m2. On baseline echocardiography, the average Vmax was 1.61 ± 0.45 m/s, and the average ΔVmax was 0.05 ± 0.27 m/s/year.

Table 1 Patient characteristics in the none calcification, mild calcification, and severe calcification groups.
Full size table

Histological findings

Microcalcifications were predominantly localized in the media (Fig. 1). These tissues were divided into three groups based on the first and second tertiles, with the first tertile at 0.017% and the second tertile at 0.256%.

Fig. 1
figure 1

Histological findings of radial artery calcification in patients with end-stage kidney disease. (a) A section of the radial artery showing no calcification (0%), (b) mild calcification (0.16%), and (c) severe calcification (9.69%). Microcalcifications were predominantly localized in the media.

Full size image

Predictors of ΔVmax: univariate and multivariate analyses

In the univariate analysis of clinical characteristics, cardiovascular risk factors, laboratory data, baseline Vmax, and RAC level, RAC level (adjusted mean of ΔVmax for severe group: 0.16 m/s/y, 95% confidence interval [CI] 0.07–0.25 vs. for mild group: 0.02 m/s/y, 95% CI − 0.07–0.11 vs. for none group: − 0.03 m/s/y, 95% CI − 0.13–0.06; p = 0.01), baseline Vmax (standardized coefficients [β]: 0.21, 95% CI 0.09–0.33, p < 0.001), and age (β: 0.00, 95% CI 0.00–0.01, p = 0.02) were significantly associated with the increase in Vmax (Table 2). In the multivariate analysis of RAC level, baseline Vmax, age, and sex, RAC level (adjusted mean of ΔVmax for severe group: 0.14 m/s/y, 95% CI 0.04–0.23 vs. for mild group: 0.02 m/s/y, 95% CI − 0.07–0.11 vs. for none group: − 0.04 m/s/y, 95% CI − 0.13–0.05; p = 0.03) and baseline Vmax (β: 0.14, 95% CI 0.01–0.26, p = 0.03) were independent factors contributing to the increase in Vmax (Table 2). The relationship between ΔVmax and RAC level is shown in Fig. 2. Higher levels of RAC tended to result in greater increases in Vmax. Conversely, diabetes, serum phosphate, calcium, calcium–phosphate product, and parathyroid hormone levels were not associated with an increase in Vmax.

Table 2 Univariate and multivariable analyses showing factors associated with an increase in Vmax.
Full size table
Fig. 2
figure 2

Relationship between ΔVmax and level of radial artery calcification. Higher levels of radial artery calcification were associated with an increase in Vmax.

Full size image

Factors associated with baseline Vmax

In the univariate analysis of clinical characteristics, RAC level (adjusted mean of baseline Vmax for severe group: 1.79 m/s, 95% CI 1.64–1.94 vs. for mild group: 1.48 m/s, 95% CI 1.33–1.63 vs. for none group: 1.56 m/s, 95% CI 1.40–1.71; p = 0.02) and age (β: 0.01, 95% CI 0.00–0.02, p = 0.01) were significantly associated with higher baseline Vmax (Table 3). In the multivariate analysis of RAC level, age, hemoglobin (Hb), phosphate, and low-density lipoprotein cholesterol, age (β: 0.01, 95% CI 0.00–0.02, p = 0.01) was an independent factor contributing to higher baseline Vmax. Although not statistically significant, higher rates of RAC tended to be associated with higher baseline Vmax (adjusted mean of baseline Vmax for severe group: 1.76 m/s, 95% CI 1.61–1.91 vs. for mild group: 1.51 m/s, 95% CI 1.36–1.65 vs. for none group: 1.56 m/s, 95% CI 1.41–1.71; p = 0.05; Table 3).

Table 3 Univariate and multivariate analysis of patient characteristics contributing to baseline Vmax.
Full size table

Factors associated with RAC level

In the univariate ordinal logistic regression analysis, only Hb level tended to be associated with RAC level (odds ratio: 0.74, 95% CI 0.54–1.01, p = 0.07). Even after adjustment for C-reactive protein level, corrected calcium level, and history of dyslipidemia, Hb level tended to be associated with RAC level (adjusted odds ratio: 0.75, 95% CI 0.54–1.03, p = 0.08), but there were no associations with all variables (Table 4).

Table 4 Univariate and multivariate ordinal logistic regression analysis of patient characteristics contributing to radial artery calcification level.
Full size table

Discussion

Our study revealed that elevated RAC level at the initiation of HD were associated with an increase in Vmax. To the best of our knowledge, this is the first report to establish a link between RAC level and the progression of stenotic changes of aortic valve in patients with ESKD. Although many cases in this study did not meet the diagnostic criteria for AS, the significant increase in Vmax may serve as an indirect predictor of its future onset and progression. The results suggest that patients with ESKD and high RAC level are at a greater risk of AS development and progression. Therefore, these findings may help identify this high-risk group and provide targeted healthcare interventions. AS is a progressive disease with varying rates of progression among patients14, particularly dialysis patients. There is no consensus on the factors influencing AS progression in this population. Some studies suggest that annual echocardiography may be beneficial for monitoring moderate or asymptomatic AS in patients with ESKD15,16. However, current guidelines recommending fixed intervals for echocardiography based on population averages of AS17 may lead to unnecessary tests for slow-progressing patients and overlook significant progression in rapid-progressing patients18. The findings of our study may aid in administering appropriate healthcare strategies for patients with AS.

The most accurate method for evaluating vascular calcification (VC) is through histological examination of arterial specimens. While some studies have reported histological examinations of VC using rodent models like rats, there are limited reports on histological evaluation using human arterial tissues. Obtaining human artery tissues is typically challenging, but it can be easily done during AVF surgery19, allowing for the pathological evaluation of human arterial tissues.

It has been reported that RAC may be frequently associated with calcific coronary plaques20. In addition, there has been a correlation between coronary artery calcification and cardiac valve calcific deposits21. Aortic valve calcification is a precursor of AS22 and is involved in its development and progression. The risk of aortic valve calcification is linked to the baseline Agatston score of aortic valves23. While aortic valve calcification was traditionally attributed to time-dependent valve leaflet wear and passive calcium deposition, recent research suggests it is an active, multifaceted condition involving lipoprotein deposition, chronic inflammation, osteoblastic migration of valve interstitial cells (VICs), and active valve apex calcification24, in addition to passive processes. VC is also considered an active, regulated pathological process similar to osteogenesis, involving a phenotypic transition of vascular smooth muscle cells (VSMCs)25. Valvular calcification is common in CKD and is closely associated with intimal arterial disease findings26. VICs, rather than VSMCs, populate the aortic valve. The pathogenesis of AS progression is distinct from that of atherosclerosis27. However, valvular calcification and VC may share common risk factors and pathogenetic features28. Our findings suggest that VC, common in CKD, and AS (aortic valve calcification) progression may be related, supporting this hypothesis.

It has been reported that baseline Vmax is a predictor of AS progression29. Another study indicated that calcified aortic valves and diabetes mellitus have been linked to rapid AS progression30. Some studies have also shown that serum phosphate, calcium, calcium–phosphate product, and parathyroid hormone levels can predict AS progression31,32. In addition, recent research suggests that warfarin use may be linked to AS progression33. However, conflicting reports exist, with some studies showing no correlation between these factors and AS progression in dialysis patients. The exact cause of the accelerated disease progression in this population remains unclear and is likely multifactorial9. In our study, as previously reported, it was suggested that baseline Vmax was related to AS progression, while factors such as diabetes mellitus, serum phosphate, calcium, and warfarin use were not found to be associated.

Aortic valve calcification increases as the estimated glomerular filtration rate decreases34. It has been reported that severe calcification in patients with ESKD actually begins long before renal insufficiency is severe enough to require dialysis35. The rate of progression of AS in dialysis patients is higher than in the general population6. In addition, decreased renal function may be associated with AS progression in nondialysis patients36. As just described, the degree of renal dysfunction is closely related to aortic valve calcification and the subsequent progression of AS. This study only included patients at the time of dialysis initiation; thus, we could study the progression of stenotic changes of aortic valve at the same level of renal function and duration of dialysis.

Currently, there are no pharmacotherapies that directly prevent the onset or progression of AS. In Japan, transcatheter aortic valve implantation for severe AS has been covered by insurance, even for dialysis patients, since February 2021. Therefore, surgical or transcatheter aortic valve implantation procedures are the only effective treatment options for severe AS in dialysis patients37. Early aortic valve replacement significantly improves 5-year survival compared to conservative treatment, even in dialysis patients38. Therefore, early detection of the onset and progression of AS and prompt treatment can improve the prognosis of these patients. For dialysis patients, there are no established standards regarding the intervals for echocardiography. Currently, the interval between echocardiography examinations after the initiation of dialysis varies from facility to facility. Patients with RAC may require more frequent follow-up echocardiography exams for the early detection of AS.

VC is commonly observed in patients with CKD and is often accompanied by abnormalities in phosphate and iron metabolism25,39. It has been reported that VC can develop in young adults with ESKD even in the absence of traditional cardiovascular risk factors such as hypertension or dyslipidemia40. The mechanism of VC in CKD differs from classical arteriosclerosis (atherosclerosis). In this study, no significant factors contributing to arterial calcification were identified, including anemia, which is associated with CKD-MBD and classical arteriosclerosis risk factors. Anemia can lead to peripheral hypoxia, which in turn influences the osteogenic potential of bone cells41. It has been suggested that the hypoxia-inducible factor signaling pathway, which is the cellular response to hypoxia, potentially plays a role in the osteochondrogenic differentiation of VSMCs42. Low Hb level tended to be associated with RAC but it was not significant, thus this should be further studied.

This study has some limitations. First, the use of an AVF for HD access can impact cardiac load by increasing preload and decreasing afterload (peripheral vascular resistance)43, leading to high cardiac output and potentially elevating the transvalvular gradient in the setting of AS44. Second, the hemodynamic evaluation parameters of the aortic valve may be modified by the dialysis session36. While most assessments were conducted on nondialysis days in the majority of cases, the exact timing of these inspections was not always known. Third, because arterial specimens were harvested during AVF surgery, arterial specimens comprised only part of the artery opposed to representing cross-sections. This might affect the quantification of the RAC level. Moreover, it was difficult to collect a wide range of data on patients beyond echocardiography data after the initiation of HD, and therefore, we could not examine the impact of these data on ΔVmax. For instance, the control and treatment of CKD-MBD and anemia might have affected ΔVmax after initiation HD. However, our findings suggest that AS may progress more rapidly in cases with RAC, irrespective of the post-HD initiation scenario. Finally, because of the relatively short follow-up period and the limited sample size, progression to AS was observed in a few cases. Additionally, variability in follow-up durations across cases made the assessment of the absolute value of Vmax difficult. Therefore, evaluations were based on ΔVmax, adjusted for the follow-up interval.

We propose that an increase in Vmax may serve as an indirect predictor of AS onset and progression. However, a longer follow-up period and a larger sample size would be necessary to draw definitive conclusions regarding RAC as a predictor of AS development or progression.

In conclusion, RAC is an independent factor contributing to an increase in Vmax after HD initiation. Patients with ESKD and RAC should be aware of the potential development and progression of AS.

Methods

This study was performed as a subanalysis of the previous prospective trial. The study was approved by the Ethics Committee of Juntendo University Hospital, Tokyo, Japan (approval number 13-101) and registered at the clinical trials registry on 31/10/2013 (registration no. UMIN000012150). The study was conducted in accordance with the Declaration of Helsinki (revised in Brazil in 2013), and informed consent was obtained from all participants, as approved by the ethics committee. The primary endpoint of this study was ΔVmax after initiation HD.

Patients

A total of 150 patients with ESKD underwent primary AVF surgery between November 2013 and April 2016 at Juntendo University Hospital, Tokyo, Japan. Exclusion criteria included insufficient sample of radial artery (n = 15), lack of echocardiography data or low left ventricular ejection fraction (n = 35), history of aortic valve surgery (n = 4), bicuspid aortic valve (n = 1), and aortic and/or mitral regurgitation (n = 1). Overall, 94 patients with ESKD were enrolled in the study. Baseline data for these patients at the time of AVF surgery were obtained from the institutional database.

Histological studies

All segments of the radial artery were collected during AVF surgery and immediately fixed in 10% neutral formaldehyde. The specimens were stained using the Von Kossa method to quantify calcium, which can detect microcalcifications. Slides were immersed in a 5% silver nitrate solution to replace calcium cations with silver, turning the mineralized material black. After being placed in 5% sodium thiosulfate, the specimens were counterstained with Kernechtrot for contrast, selectively staining nuclear chromatin red and providing nonspecific background tissue staining in shades of pink.

Sections stained with Von Kossa staining were captured as 20 × color images using an Olympus microscope (VS120-L100, Olympus, Tokyo, Japan). Histomorphometry was conducted using Definiens Tissue Studio software (Definiens, Munich, Germany). The area of Von Kossa-stained positive and the total arterial tissue were measured (Fig. 3), and the percentage of calcified tissue area to the total arterial tissue area was calculated for histologic quantitative evaluation. In case arterial tissue contained large amount of loose adventitial tissue, those sections were excluded (Fig. 3).

Fig. 3
figure 3

Methods of histomorphometry of radial artery calcification using software. (a) The region enclosed in orange represents the total arterial tissue (the loose adventitial tissue was excluded). (b) The calcium-stained area is indicated by yellow area, which corresponds to the Von Kossa-stained positive area.

Full size image

Echocardiography

These patients underwent echocardiography immediately before AVF surgery, and at least one echocardiography was performed within 5 years of initiating HD. The patients were examined in a left lateral decubitus position using the LOGIQ e ultrasound system (GE Healthcare Japan, Tokyo, Japan). The exam included a comprehensive assessment of cavity and wall dimensions, ventricular and valvular function, morphologic appraisal, and pressure predictions, following international guidelines. Vmax was routinely measured using continuous-wave Doppler tracing, recorded from the window yielding the highest velocity signal. The progression between the first and last exams was expressed as a Vmax change. To account for interindividual variations in the follow-up interval, the change in peak flow velocity was normalized to a 1-year interval. The Vmax data immediately before AVF surgery was defined as baseline Vmax, and the change in Vmax per year was defined as ΔVmax.

Statistical analysis

Continuous variables were presented as mean ± standard deviation for the normally distributed data or median and interquartile ranges for the skewed data, while categorical data were presented as numbers and percentages. For the evaluation of radial artery calcification (RAC), the tissues were categorized into three groups based on the degree of RAC: none, mild, and severe RAC groups by the first and second tertiles (Fig. 1).

First, univariate analysis was conducted using simple linear regression for continuous covariates or one-factorial analysis of variance for categorical factors to identify relevant factors potentially influencing ΔVmax (factors with a p value of 0.2 or less). These factors, along with sex, were included in the multivariate analysis as covariates.

Second, a multivariate analysis was performed to investigate the impact of the parameters on ΔVmax. Multivariate analyses using analysis of covariance for categorical variables or multiple linear regression for continuous variables were conducted, and RAC level, sex, diabetes mellitus, baseline Vmax, and age were selected as covariates.

In addition, factors associated with baseline Vmax were examined using analysis of covariance for categorical variables or multiple linear regression for continuous variables, while the degree of RAC was analyzed using ordinal logistic regression.

All statistical analyses were conducted using the Windows version of JMP software (Version 17.0.0, SAS Institute, Inc., Cary, NC), with a significance level set at a p value of < 0.05.